What is Product Analytics

Product Analytics is the term applied to the automated gathering, analysis, and visualization of data. Most people first get their introduction to analytics through the popular Google Analytics, but there is a variety of analytics software specifically for tracking user actions on digital products.

It’s not just Product Managers who need to implement product analytics. Anyone involved in software development, from engineers to designers, can use data to make more informed choices.

As a Product Manager, you don’t need to be a data scientist, but you do need to be comfortable analyzing and utilizing data. It’ll benefit you not just for your current product, but throughout your entire career.

You might also be interested in: 5 Reasons Why It’s Important for Product Managers to Understand Data

Why The Numbers Matter: The Importance of Product Analytics

“Without data, you’re just another person with an opinion.”

-W. Edwards Deming

That’s true across all aspects of business, but it’s especially true for product teams. By making data-driven decisions, you improve your product and the user experience. According to Mixpanel, “studies show that companies who rely on product analytics are far more profitable than their peers.”

When companies focus on the data, and use it to make decisions, they build better relationships with their customers. When a company understands their customers better, interactions with them improve at ever level, from customer support, sales, and within the product itself.

You can’t help your users unless you know them, and creating a data-driven company culture allows for better understanding.

Being a Data-Driven Product Manager

You may have heard the term ‘data-driven Product Manager‘ thrown around a lot in recent years. But what actually makes a data-driven Product Manager?

Essentially, a data-driven Product Manager is one who doesn’t just rely on their instincts, but someone who arms themselves with as many facts as possible.

Now that you’re a data-driven Product Manager, you might be on the lookout for some quick wins! But being dedicated to data doesn’t work like that. It’s more about playing the long game. Investing your time in data is exactly that…an investment.

You might also be interested in: What is Product Data Management?

You need to build product analytics into everything you do right from the start. Being data-driven won’t make you an overnight success. It’s a long term practice that influences your product decisions and leads you to long term success.

Product Analytics Tools: The Features You Need

There are a whole host of product analytics platforms out there, as well was business intelligence platforms who offer similar features. These are the most common and, arguably, the most important ones:

A/B Testing

Sometimes we don’t know what will work until we try it! A/B testing offers you the chance to test out your hypotheses and improve your product based on the results.

This is great for you as a Product Manager. When a stakeholder ‘requests’ a feature which you disagree with, you can come back to them armed with data and say “we tried that and it didn’t work, and here’s the proof.” It can also help you to test out different versions of your product with different users, in order to understand who prefers what.

For example, A/B Testing is a strategy that Netflix implements almost religiously. It’s a powerful tool used to help understand a user’s preferences and is one of the keys to creating personalized experiences.

User Tracking

When you start figuring out exactly how users interact with your product, how long they spend on one feature, where they drop off, etc…you’ll wonder how you lived without user tracking!

While qualitative data is useful for helping you to understand how users respond to your product, they don’t always mean what they say. There’s nothing quite as insightful as tracking users in real-time when it comes to understanding user behavior.

Once you have your users tracked, you can move onto…

User Segmentation

This will help you to work out how different users interact with your product. By organizing your users into various segments, you can figure out how to best serve each group.

You might also find new opportunities for growth with a certain demographic. For example, Pinterest found, through product analytics, that men were the most underserved demographic.

The product team used this information to offer more personalized results for these users, which led to males becoming the fastest-growing demographic within their user base.

Segmentation can also help you to understand seemingly erratic user behavior. Let’s say that a portion of your users is either ignoring or completely misusing a certain feature. Segmentation and tracking allows you to see how many of those users successfully completed the on-boarding process.

Who Benefits from Product Analytics?

The best part about implementing product analytics is that almost every team can benefit from it. According to Gainsight, Customer Success can use the data to make more proactive recommendations to customers, Marketing can use it to tailor their messaging, and Sales can use it to identify the right time to contact a prospect.

In your day-to-day as a Product Manager, having access to product analytics will be most helpful when it comes to making product decisions, but also in your conversations. Data gives you a common ground to begin discussions with cross-functional teams.

While data can be interpreted in a few different ways, it can sometimes supply you with an objective truth. For example, if you see that no one is clicking on an important CTA, the objective truth is that no one is clicking on what you need them to click on!

The next step is to take this problem and form your problem statement, giving your teams a common goal, and helping you to understand what information you need from your analytics.

Learning to Ask the Right Questions: Define the Problem Statement

While Product Analytics is a powerful tool, it can’t magically do your work for you. You need to learn how to properly wield it and how to ask the right questions.

A problem well defined is a problem half-solved

— Charles Kettering

In data science, one of the key factors for asking the right questions is developing and defining the right problem statement. After all, you won’t know what questions to ask from your data if you’re not 100% what problem you’re trying to solve. For this, you’ll need to follow three basic, but critical, steps:

1: Understanding the problem

Whether the problem comes from your users or another set of stakeholders, you need to properly understand the problem. The only way to do that is through a combination of research, and empathy. This means gathering both qualitative and quantitative data. Find out how the user/stakeholder feels about the problem, as well as how they behave in response to it.

2: Make a risk vs reward assessment

Once you know what the problem is, you need to lay the groundwork for your plan on how to proceed with solving it. Analyze the potential risks and rewards of the project. So if a stakeholder is asking that a new feature be implemented, work out how much of your team’s work hours, budget, and resources will be needed to complete the project and solve the problem.

Balance this out by seeing how the best-case scenario (eg, you completely solve the problem) will benefit your product in terms of OKRs, and the bottom line.

3: Define Success

Defining the success of a project really boils down to the final part of your risk vs reward assessment. What does the best case scenario look like?

If the answer isn’t obvious, think about your company’s North Star metrics, or your team’s KPIs. If the project is worthwhile, its goals should align with either, if not both, or these.

How to Ask the Right Questions from Data

Once you have a defined problem statement, you can work on asking the right questions from your data set. And this isn’t something you have to do alone. This is the perfect opportunity for collaboration with your data scientist.

One key thing to remember when asking your questions is to remember that data science is a science. That means asking iterative questions, testing multiple hypotheses, and using data to do more than just keep your boss happy!

Think about your goals when asking your questions, using your problem statement as a guide. What do you need to know to fix the problem? What other factors could be affecting the data? How many different data sources do you need to look at?

Finding the Right Tools

As with all product management tools, the right one for you will depend on 3 major factors; your budget, your goals, and your expertise.

The #1 most used tool by development teams might not be the most user-friendly if you’re relatively new to analytics, or it might be too expensive for your small startup. Or it may be that you’re an experienced Group Product Manager who needs something much more powerful.

Most tools offer a free trial or a tour, allowing you to test before committing. Check out The Productverse for our full list of recommendations.

If you’re particularly interested in data visualization, check out this all-you-need-to-know guide, including our top 10 data visualization tools.

If you don’t know where to begin with choosing the right tool, or you consistently find yourself running into issues with your infrastructure, consider hiring a Product Ops Manager to your team.

Still got questions? Try asking our community of 70,000 PMs on Slack.