5 things every product manager should know how to measure

Using analytics to improve your product doesn’t have to be complicated.

By Matthew Edgar
November 27, 2017
Galileo thermometers Galileo thermometers (source: Tadek Kurpaski on Flickr)

As a product manager, you need to understand who your users are in order to build a great product that meets your users’ needs. That is easy to say but much harder to do. One approach to understanding your users and finding new ways to improve your product is through analytics. However, analytics tools can seem complicated to configure and even once configured, much of the data included in the reports can seem overwhelming or, worse yet, meaningless.

Using analytics to improve your product doesn’t have to be complicated. There are simple ways to start, and you don’t have to measure everything. Instead, you want to use analytics tools to measure just a few aspects of your product’s performance and your user’s interests. Even a little bit of data collected and analyzed can help you make better decisions. To help you get started, here are five questions every product manager should ask about their users and their product, along with some ideas on how to begin answering these questions.

Learn faster. Dig deeper. See farther.

Join the O'Reilly online learning platform. Get a free trial today and find answers on the fly, or master something new and useful.

Learn more

1. What words do people use to describe your product?

One of the more critical things to measure is how the people who use your product talk about whatever it is your product does. At a basic level, these words offer clues to the problems users face and the types of solutions they are seeking. This helps you decide what features to build and what improvements to make. By knowing what users care about the most, you can also get ideas on how to prioritize the long list of tasks.

Knowing how your users speak can give you even more information. By knowing the words people use, you can get insight into the minds of your users. As you study how people speak, patterns will emerge about the user’s motivations and intentions for working with products like yours. How big is this problem they are facing? Is this a systemic problem or something new? Are your users optimistic or pessimistic about the problems your product solves? Do they view your product as an essential aspect of their lives or their company’s success?

When you get inside people’s minds, you can do a better job connecting with your users. As you adjust your product’s features to align with these thoughts, your users will feel like your product “gets” them because your product will talk and think the way they talk and think.

There are numerous ways to learn how your users speak—through analyzing forums, support tickets, and more. But doing some type of word analysis of forum posts, tickets, social shares, or other similar written conversations with users can get tricky fast. A simpler way to begin is to research the keywords people type into Google related to your product and your industry. Keyword research tools aren’t just for search engine marketers—they are a vital first step for product managers to understand something about your users and what they want from your product.

2. What content or pages are most visited or are people most interested in?

Next, you need to know how people interact with the pages (or screens) within your product. By reviewing log files or analytics tools, you can review all the pages people accessed. At a basic level, this tells you how often people accessed those pages, which can help you prioritize updates. Pages accessed more often might deserve more attention or more updates. Along with this, you can also dig deeper into the analytics tools to see which pages people went to before or after a particular page, giving you a deeper understanding of how the pages within your product work together to create an experience for your users.

The biggest problem people run into when accessing page reports is that there are usually hundreds or thousands or hundreds of thousands of pages within your product. Your product might generate dozens of unique, dynamic URLs for each user based on specific configuration settings. This can be too many pages to comprehend, leaving the product team overwhelmed trying to figure out how to make sense of the report.

To make reviewing the pages accessed in your product more effective, you can utilize content grouping features available in web analytics tools. Pages related to Feature A are grouped here, support pages are grouped there, and the administrative pages are in their own group. By organizing your product’s pages in this way, you can move beyond trying to understand how people use individual pages and can instead understand how people use a whole section of pages. This makes your life simpler and helps you get to the relevant data points you need to view faster.

As you review the pages and screens people access, remember to take note of what isn’t in the reports. Too often, we get lost reviewing all the pages people visit, seeing which pages or groups of pages are most frequently viewed. But we want to find the places nobody goes as well. So, as you review the pages or groups of pages people did access, keep an eye out for what is missing, too. Maybe some pages are being ignored because users didn’t know the information contained on these pages was even available. Of course, people might know they could access pages but don’t because they aren’t interested in what those pages offer. Either way, by exploring the ignored areas, you learn something new about your product and your users.

3. What features are people most interested in?

Along with looking at the pages people access, we want to know what features are used. Many product managers already track this at a high level. For instance, you might know that people use the feature letting them search your database, but nobody is using the feature to get email alerts related to those searches. That is an important first step for measuring product use.

Beyond knowing which features people use or don’t use, we want to go deeper into the metrics to understand how each feature is used. If the feature relies on a form, what fields of the form do people use or not use? What links or buttons within a particular part of product are clicked and which are ignored? For information-rich areas of your product, are people reading the information your product provides? These questions can be answered in a variety of ways, but one of the more effective means of addressing these questions is by using event tracking in analytics tools. The good news is, setting up a basic event-tracking code requires modifying just a little bit of code. In other cases, you only need to copy in open source scripts, with no coding knowledge required. This makes it possible for even the least technical person to start using event tracking to monitor how people interact with each part of a particular feature.

Another approach is to use heatmap tools. Heatmaps create a visual report of where people are interacting or engaging with your product. By reviewing a heatmap, you can easily see what parts of the page users scrolled to and what links or buttons users clicked on the most. Like with the event tracking method, heatmaps help you understand how users engage with your product and how you could modify your product to encourage deeper engagement.

4. How much time do people spend using our product?

A key part of product usage is the time people spend using your product. You want to avoid the most common mistake people make using this metric: assuming there is a “right” amount of time people should spend. Some products, when performing at their best, will require people to spend a lot of time interacting with the various features offered. In the case of other products, people spending a lot of time interacting with features may actually indicate a problem.

To avoid this mistake, take time to consider what amount of time you would expect people to spend before you begin looking at the amount of time people spend. Is your product designed for quick or long usage sessions? This is where surveying or interviewing your users can help you set your own benchmarks on what amount of time users should spend when working with your product. From there, you can use the time metrics to determine how your users are interacting with your product.

As you measure time, keep in mind there are two numbers you’ll want to track: total and active time. Let’s go through an example. People may open up your web app, interact with it a little, then browse away to do something else while leaving your product open in a background tab. They might return later to use it some more. The time people are interacting with your product is the active time. Total time is the active time added to the time people had your product open in the background.

Knowing both of these numbers can help you gauge user behavior. Do people interact with your product and then leave, rarely (if ever) leaving your product open in the background? Or, are people almost always leaving your product open in the background while they go do something else? This begins to change how quickly you want to time people out of a session, forcing the user to log back in. Additionally, if many of your users leave your product open in a background tab only to return some time later, you may need to add reminders to help people remember where they left off when last interacting with your product.

5. How often do people use our product?

Finally, you want to know how frequently people interacted with your product. Do people log in or interact with your product multiple times per day? Or are they interacting with your product more irregularly? Here again, there is no “normal” frequency you should be working toward. Your product is unique, and the frequency of use will differ. For a product helping people complete daily tasks, you would expect to have daily product usage. But for a product helping people track something like monthly expenses, you might only see people return after a few weeks of not using the product at all.

As you think about the frequency of product usage, remember that it isn’t just a login area, app, or website where people may interact with your product. Those are the easier places to measure the frequency of use inside an analytics tool, but another place that is often forgotten is product usage within emails. When people receive emails from your product, such as alerts or notices, you need to track these interactions as well. Do people open the emails? Do they click on the links in the email? By tracking this information, you can fully capture all the ways people interact and better understand the frequency of how users interact with your product.

Final thoughts

As you begin using analytics to understand how people use your product, remember that every product is unique. Your job isn’t to measure yourself against an average product or some “ideal” standard. Rather, the process of measuring your product requires finding the right metrics that will help you understand how to make your product work better for your users. Don’t worry about measuring everything you possibly can—there is always more to measure. The five questions discussed above are intended to give you ideas on where to begin measuring your users and your product.

If you are looking for even more ideas on what to measure in your product, check out my video series, Product Management Core Skills: Using Analytics To Inform Product Design.

Post topics: Design