Custom Buckets: Bringing more easy-to-use analysis power to Mixpanel

When competition is tight, it’s common for companies to invest heavily into making their product the more “powerful” option available. For example, your microwave probably has interesting features (popcorn mode, reheating, and defrosting) that make the microwave more powerful, but do you really use them all? Even worse, those extra buttons probably turn the simple task of heating up leftovers into a complicated endeavor. This poses an important question: Is a product truly powerful if it’s too difficult to use?

Software companies face the same challenge to make powerful features user-friendly. At Mixpanel, we call this the “power and usability tradeoff,” which is a battle between compromising on power to make a feature easier to use, or jeopardizing simplicity to enhance analysis capabilities. We aim to beat this tradeoff so we can unlock powerful data insights for both experts and less experienced users.

A great example of a feature we recently shipped that exhibits both high power and usability is Custom Buckets. This feature can be leveraged to bucket existing user segments into new, more meaningful ones. Segments like Spain, Mexico, and Argentina can be easily combined into a new segment named “Spanish Speaking Countries.” In short, bucketing segments is a powerful method that unlocks a way for users to explore complex data questions with ease.

So, how do we know we hit the power-usability sweet spot for Custom Buckets? Two months after launch of the feature, we’re still seeing consistent increases in first-time users, and 72% of them have created a Custom Bucket more than once. Product teams on Mixpanel are advancing their analyses by leveraging user age data to create “generation” engagement metrics, and marketing teams are bucketing advertising spend into “social media” vs “everything else” to measure success according to the way their budgets actually operate.

In the past, users would bucket segments by leveraging one of our most powerful features to date: Custom Properties. However, adoption of this feature was low, even though we knew the feature was powerful enough to set up new segmentation from existing breakdowns. We found less than 1% of active users were adopting this feature, which caused us to face the fact that we weren’t democratizing access to powerful analysis with Custom Properties—we were only catering to our most advanced users.

Revisiting the feature under the power-usability lens is what put us on the path to develop Custom Buckets. After speaking with users, we were able to prioritize the elements of Custom Properties that were the most valuable to them, and we leaned on that information when streamlining the feature:

In the process of building and testing Custom Buckets, user feedback was equally as helpful in understanding what we got right and wrong. That paired with Mixpanel data, of course, which helped us get a clear picture of feature usage rate and profile of users engaging to tell us whether Custom Buckets was truly unlocking advanced segmenting for everyone.

The metrics we tracked helped confirm that this feature was on the right track:

Overall, Custom Buckets is an exciting and powerful feature that brings the capability of bucketing segments to all Mixpanel users, regardless of their function or experience with the product. The balance demonstrated in this feature is a core component of how we design all features at Mixpanel, and it plays a central role in our vision to build powerful analytics that everyone can trust, use, and afford.

You can learn more about Custom Buckets and how to get started using them here.

Gain insights into how best to convert, engage, and retain your users with Mixpanel’s powerful, self-serve product analytics. Try it for free.