Analytics & Data Mastery by DigitalMarketer

Course Review:

Analytics & Data Mastery by DigitalMarketer is a course about how to collect, organize, and analyze data in order to see how your business is doing and make informed decisions. If you don’t do this type of work, you’re essentially flying blind. It’s presented by Justin Rondeau who is the Director of Optimization at DM and John Grimshaw who is DM’s Senior Data Analyst.

You’ll learn things such as figuring out which blog topics are bringing you the most engagement and leads, finding out what are people saying about you online, where you should put your advertising budget to get the best results, etc.

The instructors listed different types of KPIs for every stage of the funnel since goals vary at different stages; this also simplifies identifying problems and opportunities in your business. The best part was suggesting which KPIs to focus on based on what type of company you’re running (e.g., solo-entrepreneur, service business, e-commerce companies).

There are instances when I wished the instructors explained things in more detail. Overall, it’s a really good course. And if you’re a marketer, I would say this one is essential.

Course Notes:

The following notes from Analytics & Data Mastery by DigitalMarketer are meant to be concise, reminding me of high-level concepts and not trying to recreate the whole course. This summary is basically a bunch of notes and lessons paraphrased or quoted directly from the course and does not contain my own thoughts.

---

Module 1. An Introduction to Data Analysis

• There are only 3-5 data points that you need to look at daily to assess the health of your campaign.

What Data Analysis Is (And What It Can Do for Your Business)

• True analysis is data with a plan. It’s not enough to know the numbers… you need to know what they mean.

• Analysis in four steps:

1. Ask the right questions (creating scope): If you don’t know what you want to know, you want to know what to look for. This creates context and identifies the application of your findings.
2. Know where to look (picking the data source): There is no use asking a question where you can’t find an answer. If you don’t know where to find information and make sure that it’s accurate, you won’t be able to answer your questions. The data sources will vary, and you need to pick the right one for the right job.
3. Know what to look for (compiling data): Looking at the wrong data will hurt your business.
4. Know how it applies (applying data): You need to apply the numbers to answer your questions—turn data into learnings that drive decisions.

Here’s an example of the four steps:

1. How many orders do I get a month?
2. Check your e-commerce provider—you can also look at GA, but there is likely a margin of error to consider.
3. Find your sales totals, which can be done finding the yearly total and dividing it by # of months.
4. Basic sales metric used for other calculations.

A tougher example of the four steps would be:

1. Which traffic channels had the biggest impact on sales?
2. Check your analytics provider—you’re looking for trends here, so you want to stay within your analytics provider.
3. Identify your best traffic sources—if you’re looking at GA, then this report is done for you.
4. This tells you the most successful channels and where you should allocate more budget.

• Averages lie. The more you can step in into your data and find what’s boosting conversions and what’s depressing them, the better and more detailed your decisions will be.

• The questions you come up with are based on your campaigns and business goals.

Why All Businesses Must Be Data-Driven

• A data-driven business can make decisions on more than a hunch. More often than not, what we think is going to work for our customers isn’t going to work because we’re not them, and so we need to dig deeper into how they’re interacting with our site and what makes them act by going through our data sets.

• A business without data simply can’t grow! It’ll lack scalability, detailed insights, and rely on its “best guesses”.

• There are only three ways to grow a business, and they’re not created equal:

1. Increase new sales, which requires building trust with new traffic.
2. Increase sale amount, which requires new products/services, bundles, or price points—there is more stuff in terms of production-side that has to go into increasing the average order value. If you can’t track these things, you’ll have no idea what to expand upon (e.g., if you don’t what your successful product lines are, you could start development on something that’s not going to appeal).
3. Increase sale frequency, which requires getting previous customers to come back and purchase more.

• Averages don’t tell the whole story. You want to use averages as indicators and then dig deep to find explanations and opportunities.

• Everyone can profit from doing analysis. If you are:

• Solo-entrepreneur/shop: Keep it simple, know you “need to know” metrics, and work from there.
• Medium/large lead generation companies: Break out your KPIs by department.
• Medium/large e-commerce companies: Break out your KPIs by department.

Analysis & The Funnel

• You need to use the funnel structure for analysis for a number of reasons:

• Goals vary at different stages: When someone is a brand new visitor at the awareness stage, you want to get them to know you, make them interested, and/or make them aware of their problem. At the bottom, you want them to buy your stuff. So, you need to measure what success is differently at each stage.
• Prospect location and behavior changes at each stage: The way that people interact with your business off and on the web is different depending on whether they’re at awareness, evaluation, or conversion stage. You want to know where people are and what stage they’re in, and by separating that out, you’ll have a better understanding of what’s working and what’s not—it doesn’t make sense to try to track someone’s behavior when they’re in your funnel trying to make a purchase and treat them like someone who’s brand new to your site.
• You can diagnose your business health for each stage: It simplifies identifying problems and opportunities in your business.

What Data Matters to Your Business

• Businesses have different needs. A dentistry office is going to need to track completely different things than a brand new tech startup; therefore, you need to mold your metrics to fit what your business is doing.

• If you’re a small or new business, your challenges would be things like:

• New unestablished properties: You need to build up digital assets (e.g., writing content for a blog, building a social profile where your audience hangs out). This requires a process of testing, honing, and figuring out what people respond to (e.g., some type of content that makes them purchase).
• Limited budget: You’ll have to be extra cost-conscious in the early stages in terms of money, time, and energy. You want to make sure to make the most out of every dollar you spend—turn $1 into $2—otherwise, you won’t be around in a year.
• Limited time: You won’t have a full-time employee dedicated to analysis.

Whereas your business goals would be things like:

• High immediate profitability: Most businesses fail within the first year, so you’ll need to start generating revenue very quickly. Everything you’ll be doing needs to ROI pretty much immediately.
• Efficient data processes: You need to be able to get meaningful insights that’ll help you figure out what you should do in your business without spending too much time on numbers. You want to look at 2-3 things that really move the needle for your business.
• Build business assets: Your goal is to, through data, figure out what assets are worth spending your time and energy on. You’ll have to ensure that the data you’re tracking correlates with business growth.

• If you’re a digital, e-commerce, or membership business (i.e., you’re a business that lives entirely online), your challenges would be things like:

• Funnels are the lifeblood of the business: When you live entirely online, your sales funnel conversion rate is critical. You want to make sure you’re performing well at each stage of the funnel.
• Evaluating membership value: Knowing the value of a member is critical for estimating customer value—this is difficult when you also have many other products.
• Lots of moving parts: When you have all kinds of different properties (e.g., offers, websites, product types), it’s really important for you to be able to track what’s working and what’s not, then put your time and energy in what’s working.

Whereas your business goals would be things like:

• Master funnel conversion rates: You need to create a system to ensure your funnels are optimized.
• Accurately evaluate customer value: This goes back to the membership. You need to determine the value of a new customer very accurately. This will let you know how much you can profitably spend.
• Accurately value properties and products: This goes back to figuring out which of your assets/product lines are working well. You want to make sure you’re spending your time and energy on the stuff that’s making you the most money.

• If you’re a service or a brick & mortar business—these types of business have certain things in common such as a longer sales cycle and high-dollar offers—your challenges would be things like:

• A mix of online and offline efforts: When you’re selling bigger-ticket items or working at a local level providing services, you’ll typically have online promotions and advertising, as well as offline things. You’ll need to evaluate the marketing efforts happening online and offline, and figure out what their value is.
• Long sales cycle: People may not need the product or service you’re offering at the moment they land on your site. You have the added challenge of creating leads, then keeping them interested and aware of the business until the moment they decide to make a purchase; this means that you have to keep your costs low.
• Evaluating performance is slow: There is a big gap between when you spend the money and when you’re able to find out how your campaign performed; therefore, you need to be careful about what you’re testing and not spend too much since you can’t figure out immediately how well you’re doing. Generating a lot of new leads after spending money on a campaign doesn’t make it a success because those leads might turn out to be bad quality leads after 6 months, for example.

Whereas your business goals would be things like:

• Find leads with long-term ROI: With a long sales cycle, lead quality is critical. You don’t want to spend so much money acquiring a lead only to find out, a year later, that they’ll never purchase anything.
• Evaluate diverse lead generation strategies: You want to compare the performance of your campaign across all kinds of mediums (online/offline).
• Maintain strong relationships: You need to make sure you have a strong rapport with the leads you generate because it takes a while for them to become a customer. You want to make sure that you’re staying top-of-mind so that when it’s time for them to make a purchase, you’re there.

Data Analysis Building Blocks

• You need data! Without data, you can’t make strategic, business-building decisions; you’ll be making decisions based on ideas, hunches, and guesses, which is neither a scalable nor an effective way to grow your business.

• Data alone is not enough! Just having a spreadsheet with thousands of numbers on it is not going to help you run a better business, give your customers a better experience, or make meaningful changes to what you’re doing. You need to turn data into action through analysis.

• Analysis matters because it:

• Turns data into actionable insights: Numbers become patterns that can be exploited or improved.
• Eliminates biases and outliers: There’s going to be times in your business when you see things such as your average order value tripling or your conversion rate going through the roof, which could be the result of a mistake in your tracking, something broken on your website, a typo where you meant to type 10 but types 100, etc. Analysis ensures that odd data doesn’t skew your results.
• Contextualizes your data: You might see something that’s going poorly in your business (e.g., your AOV has dropped, visits have dropped) in the middle of the summer, for example. Contextualizing your data will give you the insight to know that every summer in your business, you see a drop due to seasonality; this lets you prepare for lower sales during that time of the year. The analysis process will give you ways to contextualize things like seasonality, competitors going in or out of the market, etc. which lets you predict what change is going to happen and help you avoid making mistakes based on incorrect assumptions or decontextualized data.

• One of the tools you’ll use to do data analysis is the analytical decision-making process, which is an 8-step flow chart that will show you each of the different things you need to do in order to take your raw numbers and turn them into meaningful actions.

• The four parts of data analysis:

1. Data: This is going to be used to identify anomalies (i.e., things that are going well or poorly in your business) and explain them.
2. A question: This is something that you want to understand or explain, presented in a formal way. For example, if your AOV has doubled, you want to know why that happened.
3. A hypothesis: A proposed explanation for your question that must be proved or disproved using the data you’ve collected.
4. Analyst’s toolkit: This is a series of ways that you can use to question and break down the numbers that you’re seeing. It helps you contextualize your data.

Module 2. Data Collection Strategies

Introduction to Google Analytics

• Here are some Google Analytics terms that are essential to know:

• Sessions: The time a visitor spends on a website. This is the uninterrupted session experience on your site where someone can hit multiple pages, but as long as they’re on your site without leaving it. A single user can have multiple sessions, and sessions will always be a much larger number than users because every user can have multiple sessions. If a user is inactive for more than 30 minutes, it defaults to a new session—inactivity will bulk this number up every now and then.
• Users: Visitors with at least one session during a given time period.
• Pageviews: An instance of a page being loaded (or reloaded) in a browser—they tell you how many pages were loaded during a time period. These will probably be the biggest number on your site.
• Bounce: This measures the percentage of site visitors viewing only one page and triggering no events—the bounce rate is more than just getting on a page and leaving. It’s a good idea to get people to trigger an event by setting an adjusted bounce rate (e.g., you can do a time-triggered event where if people are on a page for more than 15 seconds, you cut them out of the bounce category).
• Exit: This is the number of people that leave on a given page. Unlike a bounce rate where a page is actually the only one people have seen before they left and had triggered no events, an exit rate is where people have seen previous content (i.e., they’ve been on other pages or they’ve had an event occur). The exit rate isn’t inherently a big problem (e.g., you should expect a high exit rate on your thank you page).
• Events: These are user interactions with content that can be tracked independently from a web page or a screen load—it’s about what happens and how people interact with a page after it loads. This makes sense within bounce rates; if people are on a page and they click something there or interact with a tab, they’ve now interacted with the page more so than someone who’ve just gone on the page and left.
• Landing page: This is the first page a visitor views during a session; also known as the entrance page.
• Filter: This is a configuration setting that allows you to add, remove or modify your data during processing before it is displayed in your reports. This is extremely useful when you’re trying to fix your data (e.g., filter out spam bots).
• Segments: These are subsets of sessions or users that share a common attribute. Segments allow you to isolate and analyze groups of sessions or users for better analysis. These are extremely valuable since averages lie, and you need to know really what’s the depresser and what’s the booster.
• Metrics: These are the quantitative measurements of your data. Metrics in Google Analytics can be sums or ratios.
• Dimension: This is a descriptive attribute or characteristic of data. Browser, Landing Page and Campaign are all examples of default dimensions in Google Analytics.
• Conversion: In terms of analytics, this means completing an activity on your site. Conversions are achieved when site visitors complete their end goal, like downloading a pdf, making a purchase, or submitting a form.
• Goal: A configuration setting that allows you to track the valuable actions, or conversions, that happen on your site or mobile app.
• E-commerce: A type of goal that tracks transactions in your analytics dashboard.
• Campaigns: These are reports and groups your tagged traffic coming into your site.

Google Analytics Layout & Understanding

• The Audience tab tells you who your audience is—information about visitors. The Acquisition tab tells you where your audience comes from to your site. The Behavior tab gives you information about what people are doing on your site. Finally, the Conversions tab tells you whether people are taking the desired actions you want.

Tracking Site Visitors

• Tracking matters because it allows you to:

• Understand user behavior: You can see how visitors get to your site and how they use it. This will give you deep insights into what channels work best for what purpose (e.g., you might find that Facebook produces the best leads, or email is what’s actually working best to create excitement about your products).
• Segment and drill down: You want to be able to compare segments from different channels, campaigns, and creatives to see which ones work best for different purposes. You won’t only be sourcing by traffic source, you’ll be sorting by which emails work best, which promotional tactics work best, which ads work best, etc.
• Evaluate marketing performance: You’re going to go beyond just saying “This campaign did well!” to really dig deeper in order to say “This campaign perform 10% better than this other one!” so that you know which type of promotion you should use in the future.

• The way you’ll be tracking where visitors come from is by using UTM parameters. UTM parameters is a tool that lets you take any link you’re sharing and add some text to the end; this lets you figure out where people came from.

• There are four main parts to UTM parameters:

• Source: This is the referring sites and email segments that are sending the audiences there (e.g., Facebook, Twitter, LinkedIn, sites where you have banner ads, sites that are sharing links to your content, an email segment for your promotions, an email segment for your content, etc.).
• Medium: This tells you how visitors got to your site. This describes where the link that people are clicking is housed (e.g., email, banner ad, someone else’s site with an affiliate program you set up with them, etc.). Knowing the medium lets you drill down on what different things are driving clicks.
• Content: This describes the actual creative that drove the click. You’ll be able to figure which ads are performing best, which videos on YouTube are getting people excited and getting them to click on annotations, which emails you send work best, etc.
• Campaign: This describes the overarching promotion or strategy you’re leveraging. It lets you know what you’re doing in terms of email, Facebook, other people’s websites, and all the different things you’re using, all under the same umbrella of the UTM campaign; after that specific promotion is over, you can go back and find out which channels drove the most visits, which channels drove the highest converting visits, etc.

Pulling CRM & E-commerce Data

• There are formulas that you’ll be using all the time to do your analysis, which will require knowing certain metrics. Some of these metrics include:

• Unique visits: You’re going to look at this site wide and for specific product pages.
• Number of sales: For this, you’ll be looking inside your e-commerce provider or your CRM. This is your raw number of sales for a given period or a given product, etc.
• Total revenue: This will tell you how much you made during the time period within which you’re looking.
• Total returns/refunds: This will help you figure how many times customers are unsatisfied. You’ll be able to track not only the number of refunds but the total value of the refunds.

• The formulas you’ll be using include:

• Page Conversion Rate
• Average Order Value
• Revenue Per Visitor
• Add-To-Cart Rate
• Returns & Refunds

Page Conversion Rate 

• Why: This matters because it allows you to see how individual pages are performing. Whether you’re looking at add-to-cart clicks, clicks on banners in your content, etc. your goal is to get people to take some additional action, and you’ll track that by looking at the page conversion rate.
• Where to find it: Google analytics and e-commerce provider.
• How to calculate it: Total # of sales (or of actions taken)/Total # of page visits

Average Order Value 

• Why: This tells you how much your average customer is worth to your business. It’s important to look at this because raw numbers alone don’t let you know what the average take rate is for each different step of your funnel and what the end result, as far as revenue, is for your business.
• Where to find it: Google Analytics & e-commerce provider.
• How to calculate it (if you have e-commerce tracking set up in GA, you’ll be able to automatically see this without having to manually pull): Total revenue/Total # of orders

Revenue Per Visitor 

• Why: This is great because not only it tells you how much your customers are worth but how much your site visitors are worth on average—it represents what someone visiting your site is going to spend on average.
• Where to find it: Google Analytics & e-commerce provider.
• How to calculate it: Total revenue/Total # of unique visitors (users)

Add-To-Cart Rate 

• Why: This is great if you want to look at conversions at the bottom of your funnel. It lets you know what pages are having trouble making people take the action you want, and it helps you measure your cart abandonment rate.
• Where to find it: Google Analytics.
• How to calculate it: Total # of add-to-cart clicks (or the total number of purchases or conversions for the action you want them to take)/Total # of page visits

Returns & Refunds 

• Why: This gives you deep insights into customer satisfaction, and it’s crucial for forecasting.
• Where to find it: E-commerce provider or customer service.
• How to calculate it (you can look at this for a specific time period or by a specific product): Total # of returns or refunds/Total # of sales

• There are additional metrics that don’t require a formula to calculate that you should look at:

• Buyer Frequency: This metric represents how frequently visitors come back to your site and make additional purchases. People who purchase from you are more likely to come back and make additional purchases, so being able to track buyer frequency will give you powerful insights into which customer you have are more likely to become future buyers. You can find this in your e-commerce provider or CRM.
• Leads Generated: This is an extremely important metric at the MOFU because it’s what the conversion is represented by at that stage. You’re going to look at the total number of leads generated within a time period or by different lead types. You can find this in your CRM.
• Membership Length: If you have a membership site, understanding how long people stay members on average is one of the most important things you can do for your business. You’ll be able to estimate how much you can spend to acquire a customer and make revenue projections. You can find this in your e-commerce provider or CRM.

Pulling Paid Traffic Data

• Paid traffic can do many things for your business such as:

• Interest generation: It gives you a way to put your message about a product/service in front of your target audience.
• Targeting: It has very powerful targeting options.
• Scale: It makes it easy to find new audiences to speak to.

• You’ll be using some base metrics from paid traffic that you’ll then use in your dashboards:

• Paid Traffic Spend (how much you spend on each campaign): You can look at this on a total level for whatever channel (e.g., FB, Google AdWords, LinkedIn), or you can look at it at a campaign level. It’s recommended that you look at both.
• Paid Traffic Revenue (how much you make from each campaign).
• Retargeting List Size: This is an advanced tactic, but it’s a powerful way to track how your business is growing. You’ll create lists that allow you to segment your audience and track how many visitors you’re getting, how many visitors are interested in a specific topic on your site, how many people are moving down the funnel, etc.

• The formulas you’ll be using include:

• Paid Traffic ROI
• Paid Traffic EPC (Earnings Per Click)

Paid Traffic ROI 

• Why: You have to diagnose the health of your paid traffic campaigns—you want to be ROI positive. Over time, you’ll likely to see that your paid traffic campaigns’ ROI starts to dip; this is because your ads are fatiguing (i.e., the audience your targeting has seen them too many times, the ones who are interested have opted-in, and your revenue starts to drop).
• Where to find it: Paid traffic platform & e-commerce provider.
• How to calculate it: (Paid traffic revenue – Paid traffic spend)/Paid traffic spend

Paid Traffic EPC (Earnings Per Click) 

• Why: This is important because it tells you how much you can pay for clicks on your campaigns.
• Where to find it: Paid traffic platform & e-commerce provider.
• How to calculate it: Paid traffic revenue/Paid traffic clicks

• You’ll need additional paid traffic data such as retargeting lists. A retargeting list is basically a way that you can track how people have interacted with your site (e.g., you can go in and look at people who have visited your home page, but who’ve never visited a product page). Facebook and AdWords offer this capability, as well as some site popup programs where you can set the rules to be (everyone who’s visited a specific URL – everyone who’s not visited another URL).

• Retargeting lists size is a powerful way to figure out where the interest lies on your website and what kind of opportunity you’ve got (e.g., you might notice that the size of your FB Ad Templates blog post list is 180.000 people, whereas the size of your Blogging Templates blog post list is 100.000 people). In the previous example, the blogging audience is probably converting a little better, but there’s going to be less opportunity out there about your FB programs—by speaking to the larger audience, you can start to push people down the funnel.

• The reason you want to track your retargeting lists size is to look at the percentages of audiences throughout the funnel. For every given funnel, you’ll be tracking how many people have not opted-in, how many have opted-in, how many have opted-in but didn’t purchase, how many have purchased but didn’t buy the next high-dollar offer, etc. You’ll be able to understand, generally speaking, where people are getting stuck in every funnel and which parts of it are working very well.

• Generally speaking, you want retargeting lists for the following reasons:

• Efficient audience tracking: They give you the ability to quantify site visitors very quickly. For example, you can jump into your paid traffic platform and figure out how many are in your FB Ad Templates funnel; you’ll get a list of all the different people who have hit certain steps but didn’t progress to the next step.
• Audience segmentation: You can create audience lists that group visitors by interest. This lets you figure where opportunities are for pushing people down the funnel and lets you know what kind of content you should be creating because people are responding to it.
• Pinpoint funnel opportunities and weaknesses: You can identify places where prospects are getting stuck and figure out ways to optimize that part of the funnel or figure out if that’s not a good funnel for you.

Creating and Naming Retargeting Lists

• The best way to take advantage of retargeting lists is by using a careful naming structure so that when you choose to go inside of your paid traffic platform, you can do a quick search and quickly diagnose everything.

• A great way to name your audiences is to use the term Media (i.e., something that you’re using for cold traffic). Media encompasses all the different ways to bring new visitors to your site, turn them into customers quickly by offering low-dollar offers, etc. as well as the term Monetization (i.e., something you’re using for leads and customers)—you can also use Cold, Warm, or Hot. You’ll then use a hyphen and name the specific offer (e.g., lead magnet, execution plan, product, up-sell). Lastly, you’ll use a short description to denote a particular step in the funnel (e.g., Lead Magnet Visitors/No Opt-in, Lead Magnet Opt-ins/No EP Purchase, Cart Visitors/No Purchase, Purchase/No Up-sell).

For example:

• Media – Facebook Ad Templates – A: Facebook Blog Traffic
• Media – Facebook Ad Templates – B: Lead Magnet Visitors/No Opt-in
• Media – Facebook Ad Templates – C: Lead Magnet Opt-ins/No EP Purchase
• Media – Facebook Ad Templates – D: Cart Visitors/No Purchase
• Media – Facebook Ad Templates – E: Purchase/No Up-sell
• Media – Facebook Ad Templates – F: Up-sell and Exit

Starting with Data

• When it comes to data, you want to collect first and analyze second. If you have a pre-conceived notion of how you want the data to look like, it will actually start looking like that.

• You must avoid biases by:

• Pull your raw numbers.
• Input them into your dashboard.
• Analyze after you have everything.

• If you see a trend, don’t stop pulling the numbers! You need to keep pulling data in order to see the whole story because, otherwise, you’ll start seeing trends everywhere and start connecting them to other ones where you might have big issues with correlation and causation. We’re very good at finding patterns and wired to connect trends to causation.

• Just because you start seeing a trend, doesn’t mean that that’s the end all be all. You have to really dig into numbers and your key metrics.

Suppose that in Q1, you launched several discount promotions, and after looking at the success reports you saw that your sales went up each month of that quarter. You can’t just stop at this trend—keep doing this and have sales rising forever—you’ll need to look at things like your AOV and revenue gains for those months.

• When people are looking at their data, they have a tendency to dig deeper and react only to bad things (e.g., not getting enough sales, low site visits, low revenue per visitor) while taking good things at face value. It is far more dangerous to take good trends at face value because you might, for example, allocate budget somewhere that’s unfounded and set unrealistic expectations for yourself, which is incredibly wasteful—good numbers could be just as bad or worse.

• You always want to contextualize data, and never make assumptions without the whole picture.

• Anytime you see an odd figure, you need to apply Twyman’s Law, which states that any figure that looks interesting or different is usually wrong.

• You absolutely need to set baselines because if you don’t know what’s normal, you can’t start seeing what’s abnormal! This will take about 30, 60, to 90 days to do.

• Data without context is worthless. And overhyped numbers are overhyped—is you see an unaccounted for spike or dip, start digging around and you’ll find something.

Finding Outliers with Math

• You need to be able to identify outliers in your data. An outlier is a data point on a graph or in a set of results that is very much bigger or smaller than the next nearest data point.

• You could have a positive outlier or a negative outlier. When you have a positive outlier, it creates unrealistic performance expectations, which will make your team or boss really unhappy since they can’t see what they expect to see. When you have a negative outlier, it creates unnecessary fear about potential outcomes because when you keep seeing things that are outside the norm at the low end, you’re going to be worried about trying anything.

• When you do see an outlier, just omit it!

• You can set a baseline in three steps:

1. Gather data for a set time period: Avoid any sort of ad hoc analysis. You want to pick a period such as 30 days where your data is trustworthy. You also want to make sure you’re not pulling data at times where you’re doing a lot of tweaking and testing—for DM, it’s from late October to the first week of November—since the data isn’t trustworthy.
2. Calculate the average: Consider all relevant campaign metrics.
3. Use the average as your baseline: If you run a campaign and see drastic differences, evaluate if it’s an outlier or a new trend by gathering more data.

• Avoid making decisions based on real-time data!

Module 3. Top of Funnel Analytics

Top of Funnel Goals & KPIs

• If you’re not paying attention to every stage of the funnel—in terms of metrics—you actually might miss the cause of a gain or a loss due to ripple effects.

Suppose that most people at the BOFU focus on sales metrics, a company that’s only focused on sales metrics and not making sure that they have a healthy amount of awareness and traffic coming in at the top might start seeing sales slip; similarly, if they have a bad experience at the MOFU and they’re not paying attention to that area, they won’t be able to find the problem at the BOFU.

• At the TOFU, the ultimate goal is awareness. You want to make sure that you have enough people becoming aware of your brand.

• Key metrics are metrics that dictate whether or not you need to evaluate a problem or celebrate a win (e.g., if your new visitor count drops down, you want to look into that). These are the indicator metrics that you report on daily and have baselines for.

• Deep-dive or drill-down metrics are metrics that you use to diagnose what you see happening with your core/key metrics (i.e., if you see changes with your key metrics, you use these to figure out what’s working and what’s not) and make improvements in your business. These are not things that you need to look at daily—it would be too overwhelming.

• The TOFU key metrics are:

• New visitor count: You need a steady stream of new visitors to grow your business.
• Direct visit count: The people who know your brand. Often, your most profitable organic traffic source.
• Retargeting count: This is purely for segmentation purposes.

New Visitors

• You can find your New Visitors Count in GA. You’ll go to Audience, then Behavior, then New vs Returning—make sure you switch to New Users.

Direct Visitors

• You can find your Direct Visit Count GA. You’ll go to Acquisition, then All Traffic, then Channels, then you’ll click on Direct, then you’ll use the New Users number.

Total Visits

• You can find your Total Visits GA. You’ll go to Audience, then Overview, then you’ll take the Users number (not Sessions).

Top of Funnel Deep-Dive Metrics

• If you see something funky happening, those are your indicators to start diving deeper and digging into the data itself to find the problems.

• The TOFU deep-dive metrics are:

• Channel split (content and home): How people get to your content. You want to know what sources people come from and what the best-performing (e.g., in terms of revenue) sources are, etc.
• Bounce rate: How many people are leaving your content—people that leave and don’t interact with the content itself. You want to see if the bounces are actual bounces, and what’s happening from a message-match perspective (e.g., there might be something wrong from an ad scent problem).
• Branded search volume: How many people come via branded search.
• Share of search: How big a role search plays. You want to see all the different sources that are sending people to your site—you want to know how much does organic play, how much does paid play, and how much is that in relation to the other channels.
• Total pixeled audience: How many people you can retarget.
• Visits: How many people visit your site. This is really a base metric you need to see different percentages between bounce rates, etc.

Channel Splits

• In order to find the numbers for this, you’ll go to GA, Acquisition, All Traffic, then Channels—you can also go to Overview to get the graph for this. Doing this analysis will give you some useful bits about how each channel is functioning and where the problems are; so if you see a drop in new visitors, this will allow you to figure why.

Bounce Rate

• One of the reasons you want to dig into bounce rates is to figure out if people are hitting your site and just leaving, which could be caused by a message mismatch.

• You can find your bounce rate report in Audience, Overview. You can also dive deeper by going to Behavior, Site Content, then All Pages. If you see something that at 100% then it’s not really useful data; or if you see something at 0%, then it’s likely that something is broken.

Branded Search Volume

• This is more of an optional metric to look at. A great tool you can use to dig into this is SEMrush and Google Search Console.

Top of Funnel Pixeling

• The Total Pixeled Audience (whether on Facebook or AdWords) is basically tracking how many people visited your site in some number of days, so you’re looking at the entire aggregate of people who’ve come to your site and got pixeled, which is a great way to get a sense of how many unique users you’ve had for a given period of time—at DM, they track this for 180 days. The way you can do this is by creating a custom audience of people who’ve visited your entire website for a given period.

• Pixeled Segments is basically the number of people who’ve visited a particular page(s) at the very top of the funnel but not others (e.g., people who’ve visited a blog post about a certain topic but not the lead magnet page). At DM, they just do a search for (A:) since that’s how they name their segments (e.g., Media – Facebook Ad Templates – A: Facebook Blog Traffic).

• Tracking pixeled segments over time is a great way to track growth and to see how many visitors are coming and reading content but not converting. If you’re doing retargeting, tracking pixeled segments is incredibly useful to figure out where you need to be spending your time and energy as far as ad creation goes.

Visits

• You can find your Visits by going to GA, Audience, Overview, then take the Users (Total Visitors) number. This will help you figure out whether people are coming to your site. You’ll then want to look at your New Users (New Visitors), if you see a drop in Total Visitors but not in New Visitors, then you know it’s your returning visitors that aren’t coming back, which means that you have a much larger problem than something that just has to do with increasing budget for paid ads—you’re having a problem with keeping people interested in your brand.

Applying Top of Funnel Metrics

• Your TOFU goals are as follows:

• Awareness: You want to reach new audiences, make them aware of your business and offerings, make them aware of the reasons why they need what you offer, and just start to connect with them in general.
• Branding initiatives: This ties in with awareness—you want to spread your company’s name and mission.
• Segmentation: This is a great time in the funnel to start breaking down your audience and figure out what they’re interested in so that you can start talking more about it. You want to categorize new visitors by interest.

• Each type of business will have slightly different key metrics.

• If you’re a small or new business, your metrics would include:

• Retargeting lists (Key Metric): You need to track whether or not your audiences are growing and figure out what they’re interested in. Retargeting lists can be used to generally estimate how many new visitors you’ve added and how they’re interacting with your site (i.e., what they’re looking at and what they’re interested in).
• Total and new visitors (Key Metrics): This is an alternative to retargeting lists. You want to be able to figure out how many people visit your site on a weekly basis, how many of them are new visitors, and what kind of content they’re interested in.
• New visitors by content (Drill Down Metric): This goes hand in hand with the segmentation strategy. You want to know exactly what is your most popular content with new visitors so that you can understand what interests are driving new traffic—you want to push that more on the web.

• If you’re a digital business with diverse offerings, e-commerce, or membership business (i.e., you’re a business that lives entirely online), your metrics would include:

• Retargeting lists (Key Metric): When you have a lot of diverse properties, you really want which ones are driving most of your visitors.
• Total and new visitors (Key Metrics).
• Brand search and share of search (Drill Down Metrics): Because you’re a big and more established company, you should be more interested in knowing how many people visit your site because of search. You also want to look at the share of search, which is a powerful metric since it allows you to compare what percentage of people interested in your topic are looking at your offering versus your competitors; you might be driving a lot of traffic to your site, but then realize that you only have 3% of the market, and so your scale is huge and your competitors are dominating—share gives you the ability to figure out where you sit in relation to all the other people in your industry.

• If you’re a service or a brick & mortar business—these types of business have certain things in common, such as a longer sales cycle and high-dollar offers—your metrics would include:

• Direct visitors (Key Metric): Here we’re more interested in direct site visitors rather than retargeting lists and total site visitors. Direct visitors represent the performance of your offline branding campaigns.
• Retargeting lists or total & new visitors: It’s definitely worth knowing what percentage of your visitors are new versus returning.
• New visitors by content and branded research (Drill Down Metrics): New visitors by content gives you the ability to figure out what audiences are interested in—the segmentation that you do at the TOFU is very powerful because it allows you to categorize your site visitors and figure out what they want to get. Also, knowing how many people are looking for your specific brand by googling it, lets you know what percentage of the audience is aware of the brand because of its strength and the strategies that you’ve been using offline.

Module 4. Middle of Funnel Analytics

Middle of Funnel Goals & KPIs

• At the MOFU, the ultimate goal is evaluation. You want to get customers to begin evaluating your offerings (i.e., evaluating your content to see if it’s a good fit for their needs, evaluating your brand to see if they like the company, and evaluating your offers).

• Your key metrics at the MOFU—these will let you figure out whether or not you’re succeeding at this stage of the funnel—are:

• Visitor recency: This one is critical because it tells you how quickly people are coming back to your site after visiting it the first time. If your goal is to get them to start evaluating your site and offers, you want to make sure that they’re coming back. Here, coming back means every day or every week because if it takes them two months to come back, your chances of converting them are very low.
• Banner click %: This tells you how well is your content creating interest. You want as many people to click after they read a blog post or see a page where you make an offer, etc.
• Leads generated: This is the first and most important step in making people prospects.
• Returning direct visitor count: This tells you how many people are coming to your site by typing in the URL of it into their browser and who’ve already been there before. This one is very important because it tells you how many people have your brand at the top of their mind, and it tells you how many people are looking for what you’re offering because they’re coming back to evaluate your offer.
• Retargeting lists: This, again, gives you the ability to track segments (i.e., how many people saw x but not y).

Visitor Recency

• You’re going to need to create a custom report in GA to get the data for visitor recency. You’ll name the report Visitor Recency, you’ll go with Flat Table in Report Content, you’ll choose Days Since Last Session in Dimensions and Users in Metrics, you’ll choose Exclude (User Type) in Filters then Exact (New Visitor), and you’ll save the report.

• You want to pay attention to Visitor Recency: 0 and Visitor Recency: 1-7. Make sure that when you add the numbers of these two that the percentage is high since that means that a good number of people are returning within a week, which really means that you’re really effective at getting people back to your site.

If you start to see any of those two periods (and even Visitor Recency: 8-30) dip, you’ll know that you need to be pushing more people back to the site faster, so you want to spend money on things like Facebook, consider writing more engaging content, send more emails with better content, etc.

• The more you can get people to come back to your site, the simpler it is to sell them new things, get them excited about your offerings, get them to use the products you’re selling in order to make sure they continue to be customers.

Banner Click %

• The Banner Click % lets you figure out the success rate of your different content pieces—how well your content is converting site visitors—by finding out how many Unique Events and Views you’ve had for a given period. To track this, you’ll need to create events (e.g., form submission, link clicks, etc.) in GTM.

• You’re going to need to get your Blog Visits and your Banner Click %.

• Getting your Blog Visits depends on the structure of your website’s URLs. If your blog posts include /blog/ in them, then it’s easy to pull the number of visits; you’ll go to GA, Behavior, Site Content, All Pages, and you’ll apply a filter where the page exactly matches /blog/.

If your posts don’t include /blog/ in them, you’ll need to set up an advanced filter that is based on exclusion. The filter will include these three conditions:

• Exclude —> Page —> Matching RegExp —> The RegExp will include everything that you want to cut out separated by a vertical bar. For example, you’ll type in /lp/|/course/|/podcast/|/login/|/contact-us/, etc.
• Exclude —> Page —> Exactly Matching —> Your website’s homepage URL.
• Include —> Page —> Containing —> Your website’s homepage URL.

• To get your Banner Click %, you’ll go to GA, Behavior, Events, Pages, look for Event Action (this is where the destination URL is going to live), set a filter just like you did for the pages report since you only want the URLs—Event Action include button clicks, purchases, URLs, and many other actions—and hit apply. After you hit apply, you’ll be only looking at clicks that go to things that are blog CTAs (i.e., things that you know were banner clicks):

• Include —> Event Action —> Matching RegExp —> The RegExp will include everything that you want to include (e.g., all the different landing pages that you have) separated by a vertical bar.
• Include —> Page —> containing —> /blog/ (if all of your blog posts include that) OR Matching RegExp —>  whatever slugs you want to include separated by a vertical bar. This second condition will make sure that you’re looking only at clicks on blog posts, not clicks on the entire site.

• If you notice that your Banner Click % is going down, it means that you really want to drill down at the page level and look at what the click % is for an individual page and the CTAs in it. You can do this by adding on one final filter that includes the slug of that post. For example: Include —> Page —> Containing —> customer-value-optimization. Then, you’ll need to know many people visited that article during that given period. Now, you’ll calculate the percentage.

Leads Generated

• To find the number of leads generated for a given period, you’re going to need to look in your CRM.

• You can go to the next level expanding your tracking to include leads generated by your different lead magnets—you’ll need to modify the sheet to include this.

Tracking Middle of Funnel Retargeting Lists

• In the MOFU, we care about leads, so we want to look at our retargeting lists for lead magnet visitors who didn’t opt-in, as well as our sales letter visitors (i.e., people who opted-in for the lead magnet, saw the sales page but didn’t make a purchase).

At DM, they’ll do a search for B: in Facebook, since that’s how they name their segments (e.g., Media – Facebook Ad Templates – B: Lead Magnet Visitors/No Opt-in), and they add all the numbers of each list. They’ll do another one for sales letter visitors by typing in C: (e.g., Media – Facebook Ad Templates – C: Lead Magnet Opt-ins/No EP Purchase, Media – 10 Minute Social Media Audit – C: Lead Magnet Opt-ins/No Cart).

• Since you can’t track these numbers historically, you’re going to have to check on them on a weekly basis. Also, you can break things down by looking at how many visitors each lead magnet got if you want to go more complex.

Direct Visitor Count

• Direct Visitor Count will tell you how frequently people who have been on your site before are coming back to visit again by typing in the URL of your site.

• This is really about tracking the number of returning direct visits you get to your site. To do this, you’ll go to GA, Acquisition, All Traffic, Channels, then create a segment in order to exclude new visitors since you only want returning visitors—you can call it Returning Visitors: Filter —> Sessions —> Exclude —> User Type —> contains —> New Visitor. You’ll only take the number of the Direct channel since you’re only interested in Direct Visitor Count.

Middle of Funnel Deep Dive Metrics

• The MOFU deep-dive metrics are:

• Social media followers: You’ll actually track and quantify the number of people who follow you on social media and track their growth over time. You can’t look at this historically, so you’ll have to keep an eye on it regularly.
• Social shares: This one is less about the platform specifically and more about how the content that you create is dispersed around the web by others. You’ll look at where people are sharing your stuff and how many visits you’re driving to your site from people sharing your stuff.
• # of comments: If you’re using this, then it’s a really powerful metric to track how people are engaging with your content. It’ll give you a sense of where the interest lies.
• Banner click % at post level: It is extremely important to track, for any piece of content that you create, the number of times the specific offer in it was clicked! If you see that your average has dropped, you want to go back and see which posts are having really good performance and which are not. This will let you figure out whether your offers match your content, as well as how people are engaging with your website in general.

Social Media Followers

• The easiest way to track your followers is just to go pull the numbers directly from the respective platform.

• The general principle is you want to make sure that you’re getting consistent growth on the different social channels by repeating advertising strategies or avoiding frustration points with customers that cause them to unfollow you.

Banner Click % at the Post Level

• You’ll do this if you notice that your Banner Click % is lower than you wanted. You’ll dive deeper and look at some common pages to see if you can figure out which ones are under-performing, which in turn will let you know what content you need to change the offer you’re making. This is critical since your goal is to turn site visitors into leads and leads into buyers.

• Again, you’re going to look at your Blog Pages report, which will tell you how many visitors hit a specific blog page, as well as look at your CTA Clicks report, which will tell you how many people clicked on the CTAs you have on that page.

Number of Comments

• If you’re using WordPress, this will be easy to pull. Make sure you do it on a regular basis since you can’t track it historically.

Number of Social Shares

• To track the number of social shares you get, you’ll need a tool called BuzzSumo. You can use this to figure out why some piece of content’s CTA conversion might be low and what kind of content is resonating with your audience so you can produce more of it. When you use this in conjunction with your CTA report, you can start to figure out what pieces of content are actually creating leads and which ones are not necessarily transforming int what your ultimate goal is with content.

• If you want to produce content or diagnose how people are using and experiencing content on your site, using a tool like BuzzSumo lets you understand what platforms are people sharing it on.

Applying Middle of Funnel Metrics

• Your TOFU goals are as follows:

• Evaluation: Transform visitors into leads and potential buyers. This involves a lot of evaluations on the customer’s part since they have to evaluate what they think about your business, the content you produce and they interact with, and the offers that you’re making.
• Drive return visits: This is the process that you’ll use to make visitors into leads. You’ll take people who came to your site at TOFU and convince them to come back again. Once you get people to come back, your chances of turning them into leads are way higher.
• Build social media channels: Social media channels are really important because they give you a low-pressure touchpoint with your site visitors. If someone is not quite ready to give their email address, this is a good alternative until that moment comes. It’s also a way to build social proof for your website, which helps people with evaluation.

• If you’re a small or new business, your metrics would include:

• Returning direct visitors (Key Metric): This is the best measure of people who are actually evaluating your site. You know you did a great job if more people go from being casual visitors to potential buyers.
• Leads generated (Key Metric): This is you counting how many times customers give you their contact information.
• Social media followers (Drill Down Metric): A new business doesn’t have a lot of digital assets built up. In the early stage, growing your social profiles is an important step to strengthen the offers you’re making, which is a big trust signal that must be leveraged.

• If you’re a digital business with diverse offerings, e-commerce, or membership business (i.e., you’re a business that lives primarily online), your metrics would include:

• Banner click % (Key Metric): This one is good for businesses with a lot of different kinds of content online because it tells you how many people are interested in what you offer.
• Visitor recency (Key Metric): You want to know how quickly does someone who visits your site returns. The more frequently and quickly people come back, the more likely it is that they’re evaluating what you’re offering.
• Leads generated (Key Metric): You want to know how many new leads you’re getting per week and what kind of things are turning them into leads.
• Social media followers (Drill Down Metric).

• If you’re a service or a brick & mortar business—these types of business have certain things in common, such as a longer sales cycle and high-dollar offers—your metrics would include:

• Returning direct visitors (Key Metric).
• Visitor recency (Key Metric): This matters for the same reason it does for digital business—you want to know how fast it takes people to come back. The faster you can make them come back, the more likely they’ll convert.
• Leads generated (Key Metric).
• Social media followers (Drill Down Metric).

Module 5. Bottom of Funnel Analytics

Bottom of Funnel Goals & KPIs

• The main goal at the BOFU is conversion. And the key metrics here are:

• Number of goods sold: This is about how many items are you selling.
• Retargeting and segmenting: You want to know how does this affect sales and whether your retargeting lists actually turn people into profit.
• Average order value: You want to know how much your orders are worth and whether you can get people to pay more. Remember that increasing frequency of purchase doesn’t necessarily mean increased revenue, so you want to make sure that your AOV stays up.
• Revenue per visitor:  You want to know how much is a visitor to your site worth.
• Days to conversion: Not everyone who visits your site will buy immediately, and so you want to know how long does it take to get the purchase.

Number of Goods Sold

• To get the raw number of goods sold, you’ll need to go to your e-commerce provider (e.g., Infusionsoft). You’ll put that in the Unit Types Sold in your sheet.

Retargeting & Segments

• What you’re going to look at here are audiences that have hit you cart, audiences that have made a tripwire purchase, and the audiences that have made a core offer purchase. Theoretically, you can go further down the funnel by tracking the number of people who’ve purchased profit maximizer 1, people who’ve purchased profit maximizer 2, etc. However, tracking gets a little shakier the further down the funnel you go because there are fewer people and the complexity of the rules that you set gets hard.

At DM, they do a search for the following—again, they have the option to group if they want to get more granular or sum the results they get:

• D: Cart Visitors/No Purchase —> Cart Visitors (Non-TW Buyers)
• E: Purchase/No Up-sell —> Tripwire Purchasers (Non-CO Buyers)
• F: Up-sell Purchase and Exit —> Core Offer Purchasers (Non-PM Buyers)

• Tracking this gives you a general idea of how people are progressing through the funnel and whether you have problems in your funnel(s). If you see, for example, that cart visitors have increased but the number of purchasers drops, there might be a problem with your cart conversion optimization or a technical problem.

Average Order Value

• You’re going to pull the data for this from your CRM/e-commerce provider—you might have to do the math yourself. Your AOV equals your sales totals divided by the number of orders completed (Amt sold ÷ Num sold).

• If you have E-commerce set up in GA (Conversions, E-commerce, Overview), it will calculate this for you (Revenue ÷ Transactions).

Revenue Per Visitor

• For this, you need to know the amount of revenue you’ve made and the number of visitors to your site. For the number of visitors, you’ll go to GA, Audience, Overview, and Users would be the number of visitors. For the amount of revenue, you’ll go to GA, Conversions, E-commerce, Overview, and Revenue is what you’re looking for—if you don’t trust GA, you can pull this number from your e-commerce provider/CRM.

• Besides Revenue Per Visitor, it’s also recommended that you calculate your Revenue Per Visit—you’ll use Session instead of Users. These are great ways to know how much you can expect to make for every visitor and visit to your site.

Days to Conversion

• For this, you’ll go to GA, Conversions, Multi-Channel Funnels, Time Lag, then you’ll copy the numbers to the Days to Conversion Tracking section of your sheet.

Bottom of Funnel Deep-Dive Metrics

• The BOFU deep-dive metrics are:

• Channel split (offers): You want to know how visitors got to the actual offer. We do channel splits at the top funnel level, which is looking at what channels are driving new visitors to the site. Here, we want to see what traffic is driving people to your offers and then converting them.
• Bounce rate: This will tell you how many visitors leave the offer page.
• Promo email metrics: This will show you how your internal campaigns are affecting sales.
• Individual funnel conversions: This will show you at what level (e.g., lead magnet, tripwire, core offer, up-sells, etc.) things are breaking down in your funnel.

Channel Split (Offers)

• For this, you’ll go to GA, Acquisition, All Traffic, Channels, and you can see which channel(s) is providing you with the most revenue. You can dig deeper by the actual page and campaign level (e.g., what is working within paid search). If you notice abnormally low conversions on your funnel, you’ll start digging into channels, which will tell you how channel groups impact funnel activity.

Bounces (Offers)

• For this, you’ll go to GA, Behavior, Site Content, All Pages, then you’ll set an advanced filter where you only see cart pages. For DM, the filter is set like this:

• Include —> Page —> Containing —> /secure/
• Exclude —> Page —> Containing —> oto
• Exclude —> Page —> Containing —> thank

You want to start finding which offers aren’t converting. From a checkout page perspective, the bounce rate is somewhat negligible; the exit rate is more telling because someone who has viewed a piece of content, added the product to cart but didn’t make it through the checkout stage is severely problematic. If you have a high exit rate on a cart page, you likely have a message-match issue.

• Expect the bounce rate on your cart pages to be really high.

• You can also see your bounce rate for order pages. At DM, they set the filter this way:

• Include —> Page —> Containing —> /lp/
• Exclude —> Page —> Containing —> /secure/ (if they didn’t see secure, then they almost certainly didn’t see oto or thank you)

If you have a high bounce rate on an offer page, then you want to dig into it because you likely have a product-message match between the source in that page.

• If your e-commerce provider gives you the ability to see your abandon-cart rate, you can ignore all of this stuff.

Promo Email Campaigns

• One of the ways to get this data is to go to GA, Acquisition, All Traffic, Channels, click on Email, then change the primary dimension to Campaign—you’ll have to use UTM parameters for your email campaigns to get this data. You want to take the data you get and compare it to what’s going on in your email service provider (e.g., open rates, click rates, etc.).

Funnel Evaluation (Individual Funnel Conversion Rate)

• You’ll start by looking up your lead magnet visitors for a given month by going to GA, Behavior, Site Content, All Pages, put the slug of your lead magnet in the search box (e.g., /lp/ultimate-social-media-swipe-file/), and then take out the number of Unique Pageviews.

Next, you’ll need to look up the lead magnet opt-ins. For this, you’ll need to go to your CRM and search the number of opt-ins for that particular month.

Next is the tripwire visitors for that month—assuming you don’t automatically show them your tripwire offer when they opt-in for your lead magnet. Go to GA, Behavior, Site Content, All Pages, put the slug of your tripwire in the search box (e.g., /lp/social-selling-ep), and then take out the number of Unique Pageviews. You might want to add a filter if you think the results are inflated:

• Exclude —> Previous Page Path —> Containing —> (entrance)

Next is tripwire purchases. For this, you’ll go to your CRM and look for the number of purchases of that particular tripwire for that particular month. And you’ll just continue filling out the number in your sheet.

Applying Bottom of Funnel Metrics

• Your TOFU goals are as follows:

• Conversion: You want to transform site visitors and leads into actual buyers. This is your primary goal.
• Maximize funnel performance & order value: You want to optimize your funnels in order to maximize the amount of revenue you’re getting, increase the order value, optimize the experience for the customer so that they’re more likely to complete the process from visiting your site to making a purchase.

• If you’re a small or new business, your metrics would include:

• Unit types sold (Key Metric): You want to know if you’re selling enough of the right products. This is useful because it allows you to track progress throughout the funnel. You can sell a lot of things, but if you’re not selling the right kind of units (e.g., core offers, profit maximizers), you’re never going to get the money that you need out of customers visiting your site.
• Average order value (Key Metric): This tells you how much each sale is worth. This is similar to unit types sold but it’s more tied to revenue rather than looking at places where you can optimize the experience. You want to make sure that your AOV is high enough that you can afford to drive customers with paid traffic.
• Promo email metrics (Drill Down Metric): Whereas content emails are focused on giving value (e.g., sharing blog posts with them), promotional emails are focused on monetization. You want to look at opens, clicks, and unsubscribes for these emails to figure out exactly how well they’re performing.

• Small and new business have different needs than other businesses. They need to be very lean, they need to minimize the number of things they’re tracking, they need to keep the costs down, and they need to make sure that they’re ROI-ing very quickly; therefore, this stage is incredibly important for them, they don’t have the time to wait 2, 3, 4 months for these customers they’re bringing in to turn into profitable customers. These businesses are just getting started and need all the help they can get to make it through their first year.

• If you’re a digital business with diverse offerings, e-commerce, or membership business (i.e., you’re a business that lives primarily online), your metrics would include:

• Days to conversion (Key Metric): This tells you how quickly you can turn a site visitor into a customer. You want to make sure that as many people as possible are converting as quickly as possible.
• Cumulative conversion rate (Key Metric): Since you have a lot of different properties, you’ll have many different funnels to track. It’s important to look at these on an individual level; however, to diagnose the overall health of the business (i.e., how well you’re taking a new visitor, put them into a sales funnel, and then turn them from a lead into a buyer), you just want to be able to take a quick glance and see how they’re all performing in aggregate. Knowing your aggregate funnel conversion rates should give you the information you need to figure out whether or not you can continue to push traffic, then you can always drill down and look at the individual rates once you know how the cumulative is doing.
• Revenue per visitor (Key Metric): This tells you how much every visitor is worth to you.
• Individual funnel conversion rate (Drill Down Metric): Looking at the cumulative conversion lets you get the high-level information that you need to know the general health of the business; however, when you want to do a deeper diagnosis, individual conversion rate will come in. Some of your funnels are going to be better than others, and this shows you the really high-value areas to go in and optimize your customer experience in the sales funnel.

These types of businesses need to be able to evaluate the value of each of their different properties (e.g., products/services, membership website) and figure out how much their customers are worth. They’re going to have additional long-term value, and because these are established businesses, they can pay a little bit more now to acquire the customer and know they’re going to make the revenue up later on.

• If you’re a service or a brick & mortar business—these types of business have certain things in common, such as a longer sales cycle and high-dollar offers—your metrics would include:

• Days to conversion (Key Metric): Expect this to be a lot longer since you have a longer sales cycle. You want to watch how this changes over time; if it starts to trend earlier, that’s a really good sign that means you’re doing a good job of converting people.
• Average order value (Key Metric): This one is really important since you’re spending so much on leads that take a while to convert. You need to know, roughly, how much the purchase is going to be worth to make it easy to estimate how much you can spend to drive visitors and acquire leads.
• Promo email metrics (Drill Down Metric): These emails help you convert visitors into purchasers, so you’ll look at opens, clicks, and unsubscribes to see how effectively you can generate interest in the products/services you have. They are especially important if you’re trying to create new excited visitors, turn them into customers, and then come back and hit them with high-dollar offers.

These types of business are interested in tracking the value of both online and offline media campaigns, they’re looking to estimate the value of leads because they have a long sales cycle, and they want to make sure that they’re not testing anything too crazy because it takes so long to prove whether or not their leads are going to turn into good customers.

Module 6. Retention & Monetization Analytics

Retention & Monetization Goals & KPIs

• The Ultimate goal of this stage is monetization. This is going to come in two general forms:

1. Sell our existing customers more things.
2. Try to keep our existing customers happy and satisfied with what we’re offering so that they stay with us longer, keep buying from us, and don’t refund the products.

• The retention and monetization key metrics are:

• Membership retention report: This will show you whether or not the membership program is growing.
• Traffic ROI report: This will let you diagnose the health of your paid traffic campaigns. You’ll be able to figure out whether your ads are achieving ROI or fatiguing, as well as figure out how much you can spend for someone to click on one of your ads.
• Refund report: This will let you know how customers are feeling about your products.
• Monetization funnel conversion rates: The BOFU conversion rates come in when you’re trying to create new customers. Once someone becomes a customer, you’ll start targeting them with high-dollar offers, and you’ll want to know how those funnels are performing.

Membership Retention Report

• You can find the data for this in your CRM/e-commerce provider. Your membership retention report will look at how many people you’ve added to your membership, how many people you’ve lost, what your churn rate is, etc.

Traffic ROI Report

• EPC (lifetime) is a great metric that tells you how much you can spend on a click. If you’re getting $1.25 per click, it doesn’t make sense to pay $2 per click.

• EPC (past week) lets you figure out what was the amount of money you made per click. You’re going to have to calculate this yourself by tracking the difference between your previous Total Clicks and Revenue and your current ones, then using the difference in your calculation.

• You also want to track ROI, which is called engagement decay in the tracking sheet because the ROI of almost all campaigns start to drop off as more people from the audience have opted-in or decided they’re not interested, and the creative just gets old and used-up.

• Ad engagement decay (ROI), over the week-to-week period, is the best measure of when you should turn off a campaign. Sometimes, however, you might decide that you’d run campaigns with negative ROI because you know that the LTV of each customer is much higher. For example, if you’re running a campaign that’s just sending people to a blog post to put your content in front of them, it’s not very likely that you’ll see a positive ROI, which is fine since that works within a larger traffic system where you use paid traffic to distribute content, and then retarget those content viewers to turn them into leads and customers.

• DM has a rule of thumb about ROI which states that any campaign with a 75% ROI or more is a green light campaign (i.e., turn up the spend as much as possible). Campaigns that are between -30% and 74% are yellow (i.e., they could use some optimization, and they will be left as they are without increasing or decreasing spend). Anything that is below -30% ROI is a red light (i.e., you need to turn it off).

• If you run sequential campaigns (i.e., a campaign where you’re running people to a blog post, then a campaign that retargets those people with a low-dollar offer, then a campaign that retargets those people with your core offer, etc.), then you might want to lump the cost and revenue of all campaign into one since it’s a unified effort that you’re making. Breaking them out might be more helpful because you can get a really good sense of EPC for each individual one, but if you want to really simplify it, then it’s fine to lump them together.

• You want to update the data for every campaign each week, as well as keep track of what you had the previous week.

Refund Report

• This report will look at how much is being refunded and what the total value is. It’s important to track refunds in order to know how much money you’re walking away with. The other reason why you want to keep an eye on this is that you want to know what products people refund; if people keep refunding a certain product, there’s a good chance that the offer you’re making does not match the way you’re selling it. When someone becomes a customer, you want them to stay a customer; therefore, keeping an eye on refunds and ensuring that people are satisfied with what they’re purchasing is the best way to turn people into muli-buyers instead of one-time purchasers.

• You’ll find the data for this in your CRM/e-commerce provider. You’ll pull the data (refunds and their value) for each week. When you see climbing refund count or value, you want to figure out what exactly it is that people are choosing to refund; this will help you dig in and sculpt your offers a little bit better or realize that one offer or another is just not a good fit for the market anymore.

Monetization Funnel Conversion Rate

• Keeping a vigilant eye on the conversion rates of your monetization funnels gives you a good sense of what the natural flow is from one step to the other. You’ll be able to track the performance and figure out which funnels you should run more traffic to, which funnels need tweaking, and which ones you need to get rid of.

• Conversion rate tracking at the retention & monetization stage does not include tracking lead magnet performance since people at this stage are existing customers. Also, you can exclude tracking tripwire visits and opt-ins if you run monetization funnels straight to your core offer.

• You’ll either use the Conversion Rate Tracking section (you’ll pull the data for an entire week) of your Retention & Monetization sheet or use the Promotional Funnel Tracking sheet (you’ll use this every day) for more detailed tracking.

• You can find your tripwire visitors in GA, Behavior, Site Content, All Pages, paste the slug of your tripwire in the search box (e.g., /lp/get-social-listbuilding/), and you’ll pull the number of Unique Pageviews for that week.

The number of tripwire opt-ins for that particular week can be found in your CRM. This will automatically calculate the tripwire conversion rate—at DM, they shoot for 10% or more. And you’ll continue to do the same for the rest of the steps in your funnel.

• When you’re tracking granularly (on a daily basis) and you see variations, you might be able to explain them by many things. A possible reason is the natural fatigue of the campaign (e.g., you see that conversion rates move from 10% to 9 to 8 and so on), so many people saw the offer and they already have it, which means that it’s time to change it.

• Tracking the conversion rates of your monetization funnels is incredibly important because these are for people who are already your customers (i.e. they’ve already shown you that they care about what you do and offer). You want to make sure that these monetization funnels have good conversion rates—without forgetting about the general natural fatigue of funnels—because if they’re not, then it means that you’re not providing the best possible experience for your clients.

• You might also want to keep track of unit types sold. For tripwires as an example, that would be the total of tripwire opt-ins for a particular week.

• Always make sure to double-check yourself with data; it’s never a bad idea!

Retention & Monetization Deep-Dive Metrics

• The Retention & Monetization deep-dive metrics are:

• Content email metrics: You want to figure out whether your content emails are driving engagement or not. These will include your open rates, click rates, and unsubscribe rates.
• Cohort analysis: This one is important for businesses with a membership site. Basically, this will allow you to figure out what different months or weeks, depending on what your membership program is, are performing very well and which ones are a big drop off point where customers are leaving. This will not only show you where people are leaving, but also let you estimate the value of an average customer.
• Reputation score: This is kind of your barometer for how people feel about you on social media.
• Qualitative trend report: This will look beyond just the sentiment of people social media to figure out what kind of gaps exist, what information the people you’re speaking to are looking for, and what kind of products are they looking for.
• Customer lifetime value: You want to figure out how much are different customer segments worth. This allows you to figure out which entry points are creating the best, most excited, and highest converting customers.

Content Email Metrics

• In your Email Tracking sheet, Type means whether the email you’re sending is a content email or a promotional email.

• In some ESPs, you’ll have to calculate the number of emails delivered (# Delivered) yourself since they don’t subtract the number of bounced emails from the number of emails sent (# Sent).

• When it comes to opens and clicks, you want to use the number of unique opens and unique clicks so that the numbers are not inflated.

• It’s hard to judge your email marketing performance against someone else’s because it could vary significantly by industry, by what kind of content you’re driving people to, by what kind of offers you’re making, etc. The best thing you can do is to get a benchmark of your own performance.

• When it comes to evaluating performance, you want to look at how the content emails perform together in aggregate, as well as how your promotional emails perform together in aggregate instead of comparing content to promotional stuff. Also, within each type, you want to look at the different categories; within content emails, you’ll find categories such as blog, podcast, etc. and tracking these together isn’t a good idea because it might make your overall CTR seem very high or low.

Cohort Analysis

• If you have a membership program, this is going to be a very powerful strategy for you to figure out no only where members are leaving your program, but also what is the average value of some who joins.

• Cohort analysis means breaking your members up into different cohorts (groups) based on when they joined your membership program; once you do that, you’ll figure out how long they stayed in it and what’s the average value of anyone who joined during that period.

• You can slice the data horizontally to figure out how each individual cohort performs to understand how your promotions perform, or vertically to understand what points in the membership period you should optimize in order to keep people in it longer and make sure they have a better experience.

• You might, for example, find out that there is a huge drop-off on the second payment, and if they make it through that part, they’re more likely to on the third, and the fourth, etc. In this case, you want to figure out a way to improve the customer experience before the drop-off happens.

Reputation Score

• This one is particularly important for large companies. Basically, you want to keep an eye on how much negative or positive feedback you’re getting, which you can use a tool like Mention to accomplish.

• Since this is a drill-down metric, you don’t have to track it every day. Mostly, you want to watch out for negative spikes. Anytime there is a big negative spike, something is going wrong.

Qualitative Trend Report

• For this, you’ll either use a social platform (e.g., Twitter) or you’ll use a tool like Mention. Also, it’s generally a good idea if people in your social media department keep an eye on this every day.

• The general idea is that you want to get a sense of what people are talking about. They might be talking about how some piece of content or product is missing from your catalog or from the industry in general. You also want to do this for your bigger competitors if you’re a small company—you might see people talking to them about writing or creating something.

Customer Lifetime Value

• This lets you figure out how much your different audience segments are worth and lets you estimate how much they’re going to be worth in the future.

• The goal with CLV is to dig in and figure out how much a specific subset of your audience (e.g., a group of people who downloaded a particular lead magnet) is worth over a period of time. You’ll find the data you need inside of your CRM/e-commerce provider. You’ll want to compare the performance of different lead magnets over the same period, over different periods, etc.

• Analyzing this gives you a good sense of which offers you’re making (e.g., lead magnets, tripwires) are converting people and turning them into customers that are going to generate the highest LTV.

Applying Retention & Monetization Metrics

• Your TOFU goals are as follows:

• Turn buyers into multi-buyers: You want to take people who have become your customers and sell them additional products and services. There is an existing relationship there that you can leverage, customers know who you are and they like what you’re doing, so they’re more receptive to new purchases.
• Reduce customer churn: You want to ensure that customers stay engaged and on your email list so that when you offer them other products, they’re still keeping an eye on what you’re talking about, they’re aware of what you’re doing, and they excited about it.
• Maximize membership value: You want to ensure that your service and membership customers stay on as long as possible.

• If you’re a small or new business, your metrics would include:

• Monetization funnel conversion rate (Key Metric): Once someone has already become a customer, your goal is to make them into multi-buyers, which means that you’ll need to try to sell more expensive products/services. You want to know how many customers bought higher-end offers.
• Refund total value (Key Metric): It’s critical to know whether people are satisfied with what you’re offering or not.
• Traffic ROI report (Key Metric): As a small business without much brand recognition if any, you’ll need to leverage paid traffic. You’ll need to look at your traffic ROI report to diagnose the health of the different campaigns you’re running, which will let you find out which campaigns are fatiguing and need to be turned off. By keeping an eye on this, you’ll make sure that the money you’re spending is turning into valuable dollars for your business.

• If you’re a digital business with diverse offerings, e-commerce, or membership business (i.e., you’re a business that lives entirely online), your metrics would include:

• Membership retention report (Key Metric): This is a good way to figure out whether you’re adding more members than you’re losing, whether you’re effectively monetizing the members you have, and what are the reasons that make people leave. By keeping an eye on how the membership is growing or shrinking, you can pinpoint times when you have heightened exits and entrances, and you’ll be able to turn that into actionable data to grow your membership.
• Monetization funnel conversion rate (key Metric): It’s really important to keep an eye on this because it’s how you turn existing customers into multi-buyers.
• Traffic ROI report (Key Metric): Larger businesses can definitely spend more on traffic; they can give up more on the front-end, even take a negative ROI, knowing that they’re going to make up the money later on. However, even if you’re willing to take -10% or -15% on your traffic ROI, you still want to keep an eye on the health of your campaigns and figure out when it drops below that specific barrier.
• Cohort analysis (Drill Down Metric): You’ll segment all the different members you have into groups, then figure out when they came in and when they left. Once you put this together, you’ll be able to pinpoint what months in your program are ones where people leave, as well as figure out what the average value is for the different people coming into your business, which is important in conjunction with leveraging paid traffic.

Since these businesses have many different properties, monetization will sometimes be through the form of selling people totally unrelated products from the initial one they bought. You’ll need to analyze the connection between all the different properties, and understanding which properties feed well into the other ones is a good way to leverage the connections you have.

• If you’re a service or a brick & mortar business—these types of business have certain things in common, such as a longer sales cycle and high-dollar offers—your metrics would include:

• Monetization funnel conversion rate (Key Metric): Once you made someone a buyer, the ability to sell them additional products/services is one of the best things you can leverage; this is especially important for these types of businesses because of their longer sales cycles.
• Membership retention report (Key Metric): Brick & mortar and service businesses rely on repeat business, which means they can benefit from analysis as a membership service.
• Content email metrics (Drill Down Metric): You want to make sure that your content emails are keeping your business at the top of your customers’ minds.

Module 7. Running a Data-Driven Business

Analytic Decision Making

• Once you’ve collected your data and filled out your dashboards, you want to turn that data into insights that you can use to shape your business through the analytical process.

• Analyses require:

• Asking the right questions: You need to find discrepancies in your data and start questioning what you’re seeing. You want to dig in and start asking questions like “why has my AOV dropped?”, “why am I seeing fewer new visitors come to my site in this quarter of the year?” and so on.
• Generating hypotheses: This is where you take your questions and try to come to a conclusion about why what you’re seeing might be happening. Then you’ll test and evaluate the hypotheses.
• Contextualizing data: You want to identify and recognize factors that could influence your data. There are all kinds of things that you need to think about and ask yourself before you come to a conclusion—data always lies! By contextualizing data, you’ll be able to see through the bumps that are just natural and understand real trends in your data.
• Testing and evaluating: Once you’ve assumed something, you want to see if it’s true. You want to review your drill-down metrics and relevant information to prove or disprove your hypotheses.

• Take the example of a swimwear company. During their busiest season (Q2), they saw their AOV drop 13% where it went from $42.80 to 37.20. You’ll want to use the analytical decision-making process, which is an 8-step flowchart, to help you figure out what’s wrong.

Step 1. Choose Your Funnel Stage 

• This is where you choose the place you want to be optimizing. You might, for example, feel like you’re not driving enough new leads, or you’re not driving enough people to the middle and turning them into new leads, or you’re not driving enough people to the bottom and converting them, etc. For the swimwear company, the stage would be the BOFU.

Step 2. Review Key Metrics 

• Once you’ve decided where you want to focus your efforts, you’re going to go and look at the key metrics that you created that actually let you know what’s going on. When the swimwear company were reviewing their key metrics, they saw that the AOV had dropped 13% at a time when they were expecting people to buy more; however, when they were reviewing the key metrics, they also saw that the core offer conversion rate is down 5%. Often, you’ll see key metrics in the same stage dip as something has changed.

Step 3. Formulate Your Question   

• It’s not enough to just ask a simple question. You really need to think about what you’re trying to figure out, what changes you’re seeing in your key metrics, and how you can properly question them. The question you choose to ask is what sets the stage for the hypothesis you generate and ultimately tells you whether you’ll be able to test and improve your business with that testing.

For the swimwear company, they might ask the following question: What factors are driving down conversions and order value? Pinpointing the drop in conversions will help pinpoint the order value drop.

Step 4. Contextualize with the Analyst’s Toolkit 

• You’ll need to look at what might be influencing the data (e.g., changing the promotional strategy, spending less money on paid traffic, new competitors entering the market).

For the swimwear company:

• They have no current sales promotions: During their busiest time of the year, they’re not worried about driving extra sales because they expect a lot of sales. So this can’t be influencing the AOV.
• Seasonally, this is a profitable month: The order value should actually go up, not down, during this high-demand season.
• There has been no significant change in audience volume: If they just saw fewer people coming into the site, that might be it; however, they had the same number of visitors as last year.

Step 5. Form a Hypothesis 

• This is where you take your question, which is trying to figure out what’s going on with the key metrics you’ve looked at, take the context you have, and form a hypothesis about what might be going on.

For the swimwear company, they see that one of their traffic channels is driving low-quality traffic to the site. The hypothesis could be: The drop in the quality of visitors would result in a decrease in buyer-ready visitors and sales. 

Step 6. Review Drill-Down Metrics  

• Key metrics act as flags, they indicate when you need to take a look at something. You have a baseline and you have a key metric, you see a big shift from what you’re expecting to see, and you decide that that’s a flag.

Drill-down metrics are the tools that you use to dig in and figure out what’s going on. This is where you’ll be looking at more minute things, such as what kind of traffic is coming to your site, what kind of actions they’re taking, whether people are visiting new content, who’s visiting and interacting with it, etc. You’ll look at conversions, drilling down, and seeing how each individual funnel is performing.

The swimwear company will look at the drill-down metrics in the BOFU since that’s where they’re seeing their conversions and AOV drop:

• They notice that the conversion rate is down for all traffic channels: The hypothesis has been proven wrong since all channels are under-performing in terms of conversion and not just one.
• While drilling down, they noticed that the traffic split heavily favors mobile: Mobile traffic took up an unusually high proportion of visitors (e.g. last year was 40%, but this year, it took 65%), and mobile was shown to have a lower conversion rate. They’ll now need to dive in and figure out what it is that might be dropping their performance.

Step 7. Prove or Disprove Your Hypothesis 

• The swimwear company now knows that:

• The traffic channel split did not indicate any channel was under-performing: All of them were converting roughly the same, and they were all down, so the traffic channels aren’t the source of the problem.
• The platform split showed mobile was converting lower: Once they look at the historical data, they saw that mobile had climbed 25%. Mobile traffic seems to be the problem area.

Step 8. Apply Insights 

• You now will go and look at the site through the lens of “how is mobile traffic doing?”. Are mobile visitors clicking? Are they seeing all the same stuff that the desktop visitors? and so on. The reason that the AOV and conversion rates dropped is that the mobile experience was not optimized to increase conversion. The mobile sales funnel had a page where people selected the swimsuits they wanted, but the elements of the page were overlapping and people couldn’t click on the button of the swimsuit they wanted very easily.

• In order to practice analytical decision making and analyze your business:

• Watch your key metrics: You generally want to check in on these once a week whenever you’re updating the data, and see if you’ve seen any big changes from what your baseline is. They tip you off about areas for improvement or concerning performance.
• Be prepared to be wrong: Find out that your hypothesis is incorrect in part of the process. Also, you’ll typically want to check multiple hypotheses even if your first one is right just to make sure that there isn’t some other factor influencing what’s going on.
• Stay process-oriented: The analytical decision-making process is designed to prevent data overload.

Applying the Analyst’s Toolkit

• The analyst’s toolkit is a tool you use to understand exactly what’s happening with your data—simply looking at a spreadsheet isn’t going to give you the insights you need. Think of the toolkit as a series of lenses (questions) to understand your data.

• Data doesn’t start out perfect! There will be inherent flaws in the numbers you’re pulling, even if you follow the correct pulling processes to a T, you’re not going to have perfect data—there are numerous factors that influence your data.

• Sometimes numbers lie! For example, you might decide after looking at an increased conversion rate, to spend a lot of money to drive new visitors to your website; however, when you come back a month later, you realize that something about that strategy didn’t work, and it turns out later on that there was a flaw in what you were looking at as your conversion rate. This is neither the fault of the numbers nor it is your fault; it’s an incomplete picture of what’s happening in your data.

• Don’t give up on your numbers. Instead, you want to dive into the analytical process where you ask questions to help figure out what’s actually going on with your numbers.

• The analyst’s toolkit has four lenses that you need to use:

1. Historical
2. External
3. Contextual
4. Internal

The Historical Lens 

• The broad surface questions that you’re asking here are:

• What does history tell us to expect?
• Have we seen seasonal dips (i.e., do we know that in the summer, winter, or whatever season is better for us or worse)?
• How have promotions performed in the past? You’re going to offer the same products again and again to your list—you’ll use different approaches and hooks to present them again. If you have a baseline for that, you’ll be able to figure out whether your new campaigns are performing well or not.

For example, suppose that you saw your average sessions dropped 22%. Without any context, this drop could be really concerning; however, when you actually go back and look at the historical data, you might see that you had the same dip the previous year, which is actually during the holiday season (i.e., a lot of people are less engaged with their work and less focused on business during this vacation time).

The External Lens   

• The broad surface questions that you’re asking here are:

• What changes outside our control might influence our metrics? You’ll need to look at things such whether your audiences are using things differently, the tools you’re using, the way your website is hosted (e.g., your site breaking might fall into this category), etc.
• Have referring sources changed?
• Has technology changed?
• What has changed with our competitors? This might include competitors entering the market, reducing the number of people visiting your site and making purchases or the opposite if competitors are leaving the market, etc.

For example, in 2015, more visitors moved to mobile for browsing. A higher proportion of visitors on mobile means big changes for business as usual, which will necessitate optimizing the experience on that medium.

The Internal Lens 

• This is where you look at what you’re doing inside of your own business. The broad surface questions that you’re asking here are:

• Have we made changes to our strategy?
• Have we made changes to our site?
• Have we made changes to the offer?
• Are we driving new audiences? Sometimes, new audiences don’t respond the way audiences have historically; you might see dips in conversion, higher bounce rates, etc. or you might tap into a really powerful market where you see people converting higher than usual.

For example, suppose that your pixeled site visitors (i.e., unique users who visited your site during a time period) more than doubled over four weeks. These numbers might be the result of accidentally adding a new tracking pixel to your site which will double-count site visitors.

The Contextual Lens  

• The broad surface questions that you’re asking here are:

• Am I comparing raw numbers or percentages? If you’re looking at your raw number of new visitors to your site, comparing performance from last year to this year, you’ll likely think that you’re doing great; however, when you convert those numbers into percentages where you try to figure out what percentage of people come into your website are new versus old, you might actually see that it’s dipped a little bit. You want to talk a lot in ratios because it will allow you to compare data sets of different sizes and get a better understanding of what’s actually going on.
• Are my baselines skewed by outliers? Some numbers, while they may be true, are so dramatic and heavily-weighted that you have to remove them in order to understand the general trends that you see.
• Is my data correctly pulled? It’s always worth checking when you see an odd piece of data.

For example, suppose you were tracking improvements in customer experience. Two higher volume months (e.g., the number of tickets sharply declined) means that trend actually under-represents service improvement.

• Once you apply the lenses, you’re going to want to use the insights you get to clean your data. You’re going to segment and drill-down to more specific audiences (audiences that exclude the factors that are skewing it) and you analyze those. If you wrote a blog post that went viral but the visitors are of poor quality, you want to exclude that audience in order to get a better picture of what the actual baseline is that you’re looking at. Doing this will allow you to know what the true numbers are and how the health really is, then you can go back and optimize for other segments later.

Segments & Drilling Down

• Segments give you more information about your user base, and they answer very specific questions.

• By creating segments and drilling-down, you’ll contextualize your data a lot more and really dig into where some of the holes are for a particular segment.

• You want to figure out who are the people you want to get more information about, and then start digging into how they’re using your site, where they’re doing well, and where they’re dropping off.

Analytics Trends

• In order to spot analytics trends and deal with them, you’ll need to follow certain steps:

• Step 1. Identify your Data Set
• Step 2. Identify your KPIs

Step 1. Identify your Data Set  

• With the data set, you’ll be looking at:

• Web analytics (web data): This is about what pages people are viewing, how they got there, how long they stay on the site, whether they click on anything or not, etc.
• E-commerce (sales data): This is about whether they buy something, what they’re buying, how much they’re buying, etc.
• Qualitative (user/usability data): Whereas web data and sales data show you the what, usability data shows you the why.

Step 2. Identify your KPIs 

• KPIs must measure your numbers and campaign goals—based on the area where you are in the funnel, refer back to the KPI sheet.

• KPIs come from your data sets:

• Web analytics: Page level metrics (e.g., time on site, bounce rate, users, etc.).
• E-commerce: Sales metrics (e.g., purchases, AOV, RPV, etc.).

For example, at DM, the content team measures unique pageviews (users), total CTA clicks, blog content traffic source, bounce rate, average page views.

• Web analytics trends: These tell you how your visitors are interacting with your site. Is there a trend of how they’re interacting throughout the site? Is there a trend of a drop-off?

• E-commerce trends: These reveal buyer behavior. They tell you what people are buying, how they’re buying, and how often.

• E-commerce and web data miss the why; all they tell you is that people acted a certain way or they bought. You need qualitative data to unpack this data.

• Qualitative data uses data sources like heatmaps, user surveys, and session recordings, which give your data context. These tools help you visualize and see whether people see what you’re offering or not—if you see in GA that very few people are clicking on your offer, the tools will help you with that.

• You should always remember that a trend is only an indicator! You have to contextualize the indicator by:

• Asking the right questions: Data analysis is only as good as the questions you ask.
• Have the appropriate data set: You can make data say anything! You need to make sure it says the right thing.
• Evaluate if the set is a true representation.

• When you’re looking at trends, make sure they’re not imaginary, and that they fit within the context of your scope.

Business Analyses To Start Immediately

Small & New Businesses

• If you’re a small or new business, your number one priority is to maximize immediate ROI and reach profitability. And the best place to start is by evaluating your site’s traffic sources.

• You’ll evaluate your traffic sources by looking at three areas:

• High traffic pages: This could be your homepage, your blog, etc.
• Landing pages: This is where you’re pushing new visitors that you want to convert to.
• The end of your sales funnel: This will give you insights into what sources are actually driving the most traffic and creating conversions for you.

And the key parameters you’ll use to evaluate are:

• Source: This will tell you where these audiences are coming from.
• Campaign: This will tell you which promotional efforts and strategies are actually working best.
• Content: This will help you know which ads, creatives, emails, etc. are creating the most buyers.

• For the homepage, you want to look at all the sites (you’ll sort it by Source) that are referring traffic to you. You want to figure out which businesses should you build a relationship with since they’re already driving traffic to you—leverage the existing relationships further. You also want to look at the bounce rate of each source and fix it if it’s high.

• For landing pages, you want to take each sales funnel (you’ll sort it by Campaign) and then slice and dice the traffic that’s coming there. You want to find out what efforts you’re making are working pretty well to put people on a particular page and which ones are not working so well.

By looking at the different campaigns/promotional efforts/strategies, you can figure out who’s getting to your landing page and how they’re getting there. When you understand the split, you’ll know which offers are working well with which audience, and you can continue to push them there.

Digital Business with Diverse Offerings  

• Your number one priority here is to properly value your different properties and products. And the best place to start is to determine the average value of a new membership customer. You’ll need to do a cohort analysis.

Service and Brick & Mortar Businesses 

• Your number one priority here is to figure out how to properly value your offline media and leads to improve long-term performance. You’ll start by doing using your direct visitors to dig into offline media, and you’ll evaluate your customer lifetime value by lead to figure out which leads are turning into the most profitable customers.

• You’re going to have to look at your landing page visits to see how well your offline campaigns are performing.

• For the customer lifetime value, you want to:

• Choose an old lead source (evaluate something where you’ll have enough data to approximate value): You don’t want to look at CLV as an aggregate, but as whatever the point of origin was for them—you’ll go to your CRM and search for all of the people who became leads at a specific touchpoint.
• Pull your CLV metrics: You want to look at the total amount that each individual has paid, as well as the number of transactions for the customers that were created by that particular lead generation tactic.
• Remove outliers and evaluate: Use this data to value promotional strategies and leads, as well as plan future ones.

Transitioning from Reactive to Proactive Analysis

• You need to think about your business holistically (i.e., as a whole living thing). The TOFU, MOFU, BOFU, and Retention & Monetization are all connected. Changes at one stage affect all the other stages. If you start putting more people in at the TOFU, you’ll see more people become customers at the BOFU. If you break something at the BOFU and conversion rate drops, It’s going to reduce the amount of money you have available to put people into the TOFU.

• The way to be proactive is by:

• Considering how each funnel stage influences the other: Each stage’s success or failure is based upon the performance at every other stage—no funnel stage can be viewed at a vacuum.
• Predicting the impact based on historical data: You want to look at and understand historical relationships to predict what’s going to happen. You should know that if you put a particular number of visitors into your site, it will turn into a particular number of leads, which will turn into a particular number of customer, etc. Every time you see a spike or a valley at any one stage, you know how that will ripple out through the entire funnel.
• Adjusting behavior to compensate or capitalize on changes: If you see that you haven’t put as many people into the top as you do normally, you’ll need to come up with strategies to make sure that your business is going to hit its customer and revenue goals.

• You’ll start with goals. This is a critical shift where you go from trying to figure out what the metrics tell you and how you can use them to improve things to deciding what you want to happen in the future.

• Goals matter because:

• They give you a measure of success: Starting with goals allows you to decide what success looks like.
• Allow you to reverse engineer to the starting point: Knowing your desired end result lets you know what you need your audience to do at every step.

Example

Suppose that your goal is to sell 100 core offers per month. The steps you need to take people through to reach that goal are as follows: Site visitor —> content reader —> banner clicker (you want them to see your offers and click on them) —> lead —> tripwire purchaser —> core offer purchaser.

You’ll have to start at the end by using your sales conversion rates to reverse engineer the number of banner clicks required. Suppose that your numbers look as follows:

• Lead Magnet = 37.6%
• Tripwire = 8.9%
• Core Offer = 11.4%

Since you know that you need a particular number of people to click on a banner and you know the steps that they take to get to the 100 number that you’re aiming for, you’ll simply use the following equation:

100 = 0.114 × 0.089 × 0.376 × x 

100/(0.114 × 0.089 × 0.376) = x

x = 26213

You need 26213 banner clicks in order to reach your goal of 100 core offers sold per month.

You’ll then move from the end to the MOFU. This is where the banner clicks are happening, so you want to know what your banner click % is, and then use that to determine how many unique visits to the content you need. Suppose that your banner click % is 9.07%, you’re going to use the same equation:

26213 = 0.907 × y (the number of unique visits that you need)

26213/0.907 = y

y = 289007

You need to have 289007 unique visits to your content in order to achieve your goal of selling 100 core offers per month.

Now, you’ll move to the TOFU. You want to know how frequently do new visitors become content readers. Let’s assume, for this simplified funnel that 80% of your visitors become content readers:

289007 = 0.8 × z

2989007/0.8 = z 

z = 361259

You need 361259 new visitors to come to your site every month to accomplish your goal.

Now, suppose that, for some reason, there is no way for you to drive 361259 new visitors every month to your site. In this case, you want to think about stages in the funnel you think you can optimize in order to improve overall performance and reduce the strain that you’re putting on TOFU. For example, you can work on your click banner % by doubling it, which will cut the weight you put on the TOFU in half—you’ll only need 180630 new visitors per month.

• Understand that to become a proactive business, you have to:

• Understand the analyst’s funnel is connected: The analyst’s funnel and all funnels do not exist in a vacuum—all the parts work together.
• Start with goals, then reverse engineer: Use your desired end result—this could be monetary or non-monetary—to map out what success looks like and to figure out what you need to do in each stage of the funnel.
• Track and optimize along the way: You don’t want to just start with tracking and optimizing, you want to start with what you want to accomplish, then you can optimize.

Additional Notes

• For your Pixeled Audience Report, you’ll have to do it on a regular basis (e.g., John does it every Thursday morning) since you can’t pull the data historically.