Behind the data: How Mixpanel builds governance into product

Speed and experimentation are everything in digital analytics, but only if you can trust the data. Without structured governance, dashboards get cluttered, event definitions get messy, and confidence in the data starts to erode. Suddenly, instead of debating insights, your teams debate whose numbers are correct.  

At Mixpanel, we see governance as a design principle woven into our product. Good governance means that everyone—from data analysts to business leaders—can confidently answer questions knowing the metrics are accurate and reliable.

We focus on three core pillars:

To better understand how these pillars come to life, we chatted with the team building the future of governance at Mixpanel to hear their perspective:

SP: “Our teams are structured like a triangle of execution, where each stakeholder serves different functions. The product managers are in front of the customers to get the big picture, taking time to understand their perspective and priorities. This helps us define the problem that we need to solve, so we can then shape how we want to implement it.”

JC: “We’re really focusing on making governance and data clarity more embedded across various tools within Mixpanel–and not just in Lexicon, for example.”

SM: “I'm constantly translating between what customers need and what's technically feasible at Mixpanel’s scale. I spend a lot of time understanding the friction points in our customers' workflows and their data quality demands, and then working with Sonya and Jess to turn those pain points into elegant solutions. The magic happens when we can take something complex like data quality monitoring and make it feel naturally woven into the Mixpanel experience.”

JC: “Governance is always changing. As your company grows, your teams and even roles will grow as well. It’s not a one-time thing–you can do a bunch of stuff up front to help set your teams up for success, but those things will need to evolve with you as the company grows.”

SP: “I agree with Jess and think another misconception is that data governance is just a one-person job, where they’re the only ones deciding and following the rules. It’s more like a team sport.”  

SM: “I'd add that many customers think governance means restricting access or slowing down teams with approvals and gates. In reality, good governance is an enabler. It's about creating guardrails that let teams move faster with confidence. When you have clear taxonomies, verified events, and automated quality checks, you spend less time in 'data archaeology' meetings and more time actually making decisions. Ultimately, you don’t want your team to hesitate on the data they use to drive decisions.”

JC: “The challenging thing with governance–but also all of Mixpanel–is that each customer is very different, and we’re creating a product that fits startups and their small teams, as well as larger enterprise teams. They need different things for their different stages and size of company, so building a product that is flexible enough–and also opinionated enough–is a very interesting balance.”

SM: “One of the toughest challenges is making governance visible at the right moments without being intrusive. We need to surface data quality signals when they matter-like when someone's building a report or analyzing a trend, but not create alert fatigue. It's about embedding governance into the natural workflow rather than making it a separate destination that teams have to visit. That balance of being helpful versus noisy is something we're constantly refining based on customer feedback.”

JC: “A lot of our roadmap vision was defined by customer feedback. We initially started with high-level aspirations, then we received quality input which helped turn those into real projects. For instance, some of our work shaped the direction and output for the Analysis team, who is behind the Experimentation and Feature Flag functionality—all of it defined by customer feedback.”

SP: “Another specific example is Verified Content. We had a few customers from different industries who shared the same request: the ability to tell if their data (i.e., board, properties, or events) is actually good to be used. This feedback came straight from customers and shaped the solution we released in 2025.”

SM: “One pattern we kept hearing was that governance felt like too much manual overhead. Customers wanted to establish trust in their data without duplicating effort across tools. That insight shaped our strategy around integrations: working with partners like Avo for tracking plan management, building robust data warehouse connectors, and designing with a 'govern once, apply trust everywhere' mentality. The idea is that when you verify an event or define a standard in Mixpanel, that governance should propagate—whether teams are querying in your warehouse, building in other tools, or analyzing in Mixpanel itself. We want governance to be a shared foundation, not siloed work.”

SP: “I think it’s going to be a lot more automated in the future. With governance, I feel that human-in-the-loop is still going to stay, like alerting users of data outliers, prompting user action to resolve or ignore, or detecting possible causes of data quality issues for users to address. So, it will be less monitoring—and more providing information and guidance to troubleshoot problems.”

JS: “Strong plus one to all of that. My goal with owning a lot of the automated governance work is that it will also have trickle-down and trickle-up effects, where it’ll make not just governance but all of our AI investments and Mixpanel implementation faster and easier. It’ll also help teams use and adopt Mixpanel. If we have more automated governance, people will understand their data easier and they can use Mixpanel more—versus getting stuck when they don’t know what an event is or what property to use. It will all help the entire journey.” 

SP: “I definitely agree with that—it’s like a feedback loop that feeds into itself. Let’s say… governance is a problem, you fix it and have better governance projects, and then you’re able to take advantage of more data governance tools in the project, and then it keeps looping. But if you’re off track, then all your data is off as well and the AI tools are no help to you.”

SM: “AI fundamentally changes the governance equation. When you have natural language querying, the quality of your event taxonomy becomes even more critical because the AI is only as good as the metadata it's working with. AI only accelerates that reality: garbage in, garbage out. I see a future where governance becomes more predictive: not just telling you 'this event has anomalous volume' but 'based on your implementation patterns, this event name might conflict with these three others' or 'this property definition is incomplete in ways that will limit your analysis options.' The AI becomes a governance co-pilot that helps maintain quality as you scale.”   

SP: “Turn on data volume monitoring and use the governance features! But also, planning your implementation the right way from the start.”

JC: “I’d say for larger customers (or at least what I have seen with their teams), it’s important to find a balance of tracking just the right amount of data. Sometimes, their instances can get very cluttered as they scale, like stakeholders have different naming conventions for events or they add every event into Mixpanel. Essentially, make sure you’re using what you’re tracking.”

SM: “I agree with Sonya and Jess. We've put a lot of thoughtful design into shaping these governance tools, and we want customers to actually leverage them. Beyond that, my advice is to focus on your core events and properties first—nail those with high-quality descriptions and clear ownership. Use tools like the Mixpanel Data Inspector to understand what events are being sent from each page of your application. That visibility is critical. Too many teams try to govern everything at once and get overwhelmed. Start with what matters most to your business, build that solid foundation, and expand from there.”

Governance used to mean cleaning up messes after the fact. With AI reshaping these workflows, it’s becoming more proactive—catching issues early and automating quality at scale. 

Our vision is simple: governance should accelerate your teams to move faster, not slow them down. By weaving governance natively into Mixpanel, we help you make clearer decisions with confidence and spend more time building great products—not fixing broken data.

We’re building this future together. Join the Mixpanel Community to connect with a global network of your peers, share your ideas and feedback, and help shape data governance in Mixpanel.