Frequently Asked Queries - Chapter 01 - Mixpanel
Chapter 01

Analysis trends

In this section:

in partnership with

Analysis trends

What kind of analysis should I use?

We heard (actually, we saw 😃) it loud and clear: engagement analysis (if you’re a Mixpanel user, you’ll know this as Insights) is used 3x more often than all other reports, and accounts for more than two-thirds of all analysis. Engagement analysis is also saved, added to a dashboard, shared, or exported 2x more than any other report.

Along with engagement analysis, product teams analyze their data in a variety of other ways. Here’s how their reporting breaks down in order of usage, regardless of segment, activity level, or region:

Explanation of different report types

ico-stack-insights Insights
(Engagement Analysis)

Slice and dice user behavior trends to explore them from all angles.

ico-stack-funnels Funnels
(Conversion Analysis)

Track the conversion rate of people that go from one event to the next in your sequence.

ico-stack-explore Explore
(User data)

Gain a deeper understanding of how users interact with your website or application--and who those users are.

ico-stack-retention Retention
(Retention Analysis)

Understand how long users continue to come back and find value from your product.

ico-stack-flows Flows
(User Journey Analysis)

Identify the most frequent paths taken by users to or from any event.

ico-stack-dashboards Dashboards

Dashboards aren't technically a report but an overview of all your saved and most important KPIs to track.

ico-stack-experiments Experiments
(Experimentation Analysis)

Analyze how A/B test variants 
impact your metrics.

ico-stack-impact Impact
(Launch Impact Analysis)

Measures the effects of product or marketing launches on your key metrics

ico-stack-signal Signal
(Compare Events)

Measures the association between a correlation event and a goal event and quantifies the correlation between the two.

Why is monitoring engagement so popular? That’s simple: it provides plenty of room for open-ended exploration of all the events that happen in your product. This kind of analysis lets you freely explore activity and engagement data, and it’s flexible enough that all users can easily drill into the details to find actionable insights.


“Most of the ad hoc reporting (70%)  happens in Insights, and most of our dashboards depend on saved Insights reports. The ability to break down events by user properties and event properties as being particularly useful, especially when thinking about internationalization and seeing product usage in one country versus another, or total premium conversions broken down by country.”

Jeremy Yuan Product Marketing & Analytics at mmhmm

Try it yourself

Explore Mixpanel

See how an e-commerce product uses Mixpanel to determine how many orders were placed in the US in the last 30 days, broken out by category.

Try engagement analysis with sample data

“Our most popular dashboards will have a variety of different reports in them, but three things stand out: 1) Tracking conversion with Funnels (and not just looking at the top of the funnel, but looking at how users are completing the tasks they have started); 2) Looking at how users are coming back to different parts of the product with Retention reports; and 3) Given the broad feature set we have, understanding how people are navigating between different parts of the product with Flows is super important.”

Miloš Ranđelović Head of Product Analytics at Xero

“For the data team at Chope, we’re primarily focused on search and recommendations within the ecosystem. This means we pay very close attention to the search funnel, the engagement that takes place on our search page results, and our click-through rates for the recommendation page.”

Dr. Raymond Chan Senior Data Scientist at Chope

While Data-Informed users leverage all reports, they are primarily focused on three key metrics—engagement, conversion, and retention—using reports that answer these questions at a rate of 3x more than Data-Curious users.


Ratio of reports viewed by Data-Informed vs. Data-Curious users 

It’s interesting to note that not all segments are looking at retention as closely as they are at conversion, with some companies being 4x more likely to analyze conversion compared to retention. We’ll dive into these further below. 

Industry differences

When looking by industry differences, Retail bucks the trend of favoring engagement analysis. Instead, analysts in the Retail space are drawn to conversion analysis, likely because they’re closely monitoring purchase flows. Meanwhile, Gaming relies on retention analysis more than any other industry because stickiness (and addiction 🎮) is more of a priority than in other industries.


Segment differences

Unsurprisingly, teams’ approach to data is closely related to the type of company they work for. Here’s how we define key company segments:

Digital Transformers

Companies just discovering the importance of product data. Example: An established nationwide retailer focused on diversifying its operations and revenue through digital channels.


High-growth companies that have had their idea and product validated by the market. Example: A digital marketplace that has achieved product-market fit and is growing like gangbusters.


Companies that are trying new ideas and taking risks, and have received little to no funding. Example: A company in its infancy that just announced its Series A funding on Crunchbase.

Tech Giants

Companies where data is core to their operations, regardless of their business model. Example: A household name whose app is likely on your phone.

Although engagement analysis is the most popular form of reporting for companies of all types, there are important differences in how mature and growing companies approach their data.

Startups run retention analysis more than any other segment, perhaps because it’s an important marker of product-market fit. Similarly, Scaleups and Startups are 2x more likely to find value in impact analysis (described in “other reports” above) than Tech Giants. That could be because they need to be deliberate about prioritizing developer resources and use data to identify the features most likely to have the greatest impact.

Finally, while Startups, Scaleups, and Tech Giants all run experiments at around the same rate, Startups dedicate the most time to analyzing experiments. This suggests that Startups understand the importance of making the most out of every experiment they run (indeed, their survival may depend on it).

“It’s just as important for larger enterprises to use product data to guide their business decisions as startups—both should be looking to make quick and informed decisions around product iteration. However, it’s often that successful startups are known for this practice, as the ones who don’t are the ones who fail. It’s simply too competitive and they typically don’t have the runway or cash buffer to move so slowly. Larger businesses often have horrendously wasteful processes that aren’t led by data and don’t allow for rapid iteration based on learning, though this is masked by a larger bankroll which translates into the luxury of time that startups simply don’t have.”

Janna Bastow Founder at ProdPad & Co-founder at Mind the Product

“As a small company, the stakes are higher. Time and gravity are not on your side. It’s imperative to learn fast, fail fast, make adjustments, and push forward. Unlike a big company, startups have an impossibly thin margin for error. They aren’t able to absorb mistakes or negative externalities easily. Success is made far more likely if startups are hyper-tuned to the world around them. And no set of data is more important than that which comes from customers using your product.”

Egan Montgomery Director of Go-to-Market at High Alpha

The takeaway

Engagement analysis helps you get a high-level pulse on what’s happening with your product, but for actionable insights that serve Data-Informed product development, you need to go deeper with purpose-built reports. For example, conversion analysis can help teams figure out what’s driving signups, while user journey analysis can reveal what a user did prior to adopting a feature or upgrading, and retention analysis sheds light on what keeps users coming back.

To best understand the overall impact of users’ behavior and the details that make that data actionable, teams should alternate between high-level and granular views of their data, using different report types to find the insights they need.

Resource from Mind the Product

Making Data Actionable 

Matthew Curry, Director of e-Commerce at Lovehoney, shares his experiences of making data actionable focusing on analytics and decision making in product management.

Analysis trends

How should I visualize my data?

For Data-Curious and Data-Informed users alike, creating reports is just the tip of the iceberg. Once they have the reports they need, they focus on making the information easier to consume with visualizations that turn complex information into at-a-glance insights. 

For engagement analysis, more than 45% of all visualizations are line charts that provide a quick glimpse of trends over time. Nearly 55% of teams opt for non-default visualizations like bar, table, and pie charts to better serve their unique needs and preferences.

Data-Informed users turn to bar charts more frequently than Data-Curious users, and they rely on them nearly the same amount as they do line charts. Table and pie charts are not commonly used, but Data-Curious users are more likely to use either of them.

For conversion analysis, more than 60% of all visualizations are funnel steps that display the drop-off between each step. Following funnel steps, it’s most common to see trends and time-to-convert used. Data-Informed users not only try to build and analyze the right funnel, but are also 30% more likely to track how performance changes over time.

Try it yourself

Explore Mixpanel

See how a SaaS company can take their free to paid conversion analysis a step further with six different funnel visualizations.

Try conversion analysis with sample data

The takeaway

When in doubt, use default data views—they’re the default for a reason😁. Additional visualization options in your product analytics tool exist to help you make data more easily consumable (and therefore, actionable). If the default doesn’t work for you, don’t be afraid to experiment with other ways of viewing your data.