Creating Dynamic Content Experiences

Outcome: Closed $13.93 Million gaps in ARR by allowing teams to transform their data within Mixpanel without additional cost

To build stronger customer relationships, businesses need to understand users across the entire journey. Mixpanel makes that possible by answering key questions in one place—but only if the data is in good shape.

I led the design from research to launch to reduce friction in our core product. My focus was on Content Creators—those setting up reports for their teams. By making it easier to get rich insights from existing data, I helped businesses get more value from Mixpanel without extra cost.

This solution not only improved the user experience but also reinforced Mixpanel’s position as a flexible and trustworthy analytics platform.

Team

Services

USER RESEARCH
INTERACTION DESIGN
0 → 1
1 → 2

Date

2019

ROLE

LEAD DESIGNER

Audience & Problem

Collecting data is easy—getting insight isn’t.

In small teams, tracking can adapt quickly. But at scale, data collection (engineering) and data use (product, marketing) split, slowing analysis and decision-making. Big companies add data teams to bridge the gap—but that adds overhead for everyday needs.

So we asked: What if every Content Creator (Product Manager, Analyst, Marketer) could shape their data without waiting on others?

For instance, instead of raw timestamps, imagine seeing users as “early,” “on-time,” or “late.” Custom segments like these make insights clearer, actions faster, and collaboration smoother.

Problem

  • Reduce friction between cleaning up data and getting actionable insights to teams

Goal

  • Enable Content Creators to define insights in their own terms
  • Reduce time tweaking data from days to seconds
  • Closed $13.92 Million gaps in ARR
  • Metrics to keep an eye on: DAU, interactions per session, customer satisfaction, support tickets reduction

Approach

Strategy & Kickoff

During discovery, I uncovered 2 key friction points:

  • Those who track data (engineers) are separate from those who use it (PMs, marketers, analysts), causing gaps, inconsistencies, and repeated clarifications
  • Content Creators hit a wall when they want to dive deeper into user engagement behavior and have to wait on engineer to send more detailed data to Mixpanel

As a result, Content Creators are slowed down, unable to quickly deliver meaningful Insights, Flows, Funnels, and Retention reports their teams need to make informed decisions.

I contributed to product strategy by helping my team:

  • Define a smoother workflow that aligned with how users naturally analyze data
  • Map the mental model of Content Creators, ensuring the product fit their needs and thinking
  • Prioritize key opportunities—the high-impact questions companies rely on Mixpanel to answer—guiding what we built next and why

This ensured our roadmap was both user-centered and strategically focused on delivering business value.

Milestone 1: Behavioral Analysis

To help Content Creators answer complex questions about user behavior, I ran design sessions with product managers to learn what an ideal analysis workflow looks like. They needed more advanced operations to aggregate and group.

  • Direction A: Define behaviors upfront when asking the question
  • Direction B: Define behaviors after viewing basic segmented data

In testing, Direction A aligned best with how Content Creators think—they value efficiency and see the query as the central place to ask all product questions.

A New Flexible Query: A Balancing Act

With behavior definition now built into the query, I hit a few fun design puzzles:

  • Will users understand the new actions?
  • Can they clearly interpret each element?
  • And my favorite constraint—no natural language (ask me for the keynote!)

To improve clarity, I tested several iterations. The trickiest part? Balancing interaction between the question (query) and the answer (chart).

You might ask, “Why not go vertical and spotlight the chart?” We’re headed there—but not this quarter. For now, I worked within the horizontal layout product constraints.

Visualizing the results

Asking a behavioral question is just half the battle—visualizing it clearly is the other.

Content Creators asked: Am I looking at one event? One segment? People? 


So I tested color palettes to match the new mental model.Turns out, single-color schemes didn’t work across chart types like lines or stacks. 



The winner? Categorical colors—clear, flexible and made everything click especially in Dashboards, where Consumers needed to identify patterns quickly.

Closed Beta Feedbacked

Content Creators loved the flexibility to asked more nuanced questions without having to wait days for their development teams to send extra data to Mixpanel!

But the list operations was very long even when split across two controls. In the next iterations, I focused on simplifying this to make it more intuitive and comprehensible. Eventually I landed on extended menu with visual cues on hover.

Milestone 2: Custom Segmentation

Next, I tested 2 ways to help Content Creators build custom segments:

  • Direction 1: A guided, two-step flow
  • Direction 2: Freeform input

While the guided approach was friendlier for beginners, it lacked flexibility. Since most beta users were data-savvy (often with SQL skills), I went with freeform for maximum power and control.

Sharing Content

To maintain trust with customers, it was key to give each team access to the right segments—without flooding them with too much custom content. Implementers also needed visibility into custom content in  Mixpanel’s Data Management library to stay in control during maintenance and billing conversation.

Custom Segmentation Across Mixpanel

Custom segmentation made Mixpanel reports way more valuable—teams can now explore engagement, retention, and user journeys using language that actually makes sense to them. Finally, analysis feels like it belongs to the whole company, not just the data team.

Enabling Confidence Beyond the Product

I teamed up with Brand to showcase the value of the new workflow on our Analysis page and worked with Support to highlight real customer questions Mixpanel now answers—building clarity and trust beyond the product.

Results

Impact

  • Closed $13.93 Million gaps by allowing teams to transform their data within Mixpanel without additional cost
  • Reduced data tweaking from an average of 1 week to 3 minutes
  • Interactions per session
    • 78% custom content created and saved successfully
    • 6,271 questions ran by a team had custom content (30 days)
    • 89% of reports added to dashboard with custom content

Learnings

Design

  • Flexibility isn’t a nice to have, it’s essential so Content Creators model the product into a critical tool; one that’s hard to replace, is widely adopted, and deeply embedded across teams
  • Give a preview of the possibilities via templates and more powerful possibilities in Help Docs to ensure the product doesn’t overwhelm Content Creators with too much functionality all at once.
  • Make composing a formula a delightful experience by guiding Content Creators with unique iconography; this also helped other people in the team learn the data that makes up the custom content

Leadership

  • Created a feedback board to prioritize the use cases our customers value
  • Organized design sprints to help team reframe customer problems and to generate explorations (this inspire the broader design team to do a company-wide design sprint to redesign the core product to what you see today✨)
  • Partnered with product manager to quantify the product gaps in terms of ARR to get additional team support for GA release

Next Steps

  • Allow Content Creators to get a preview of what they are building with sample data to give them confirmation they have transformed the data successfully before exiting the modal