A walkthrough of the product analytics work that transformed how a B2B SaaS company understood its users, from fragmented guesswork to a unified metrics framework that directly influenced roadmap decisions and drove measurable growth across the product portfolio.
When I joined the analytics function, three separate products, each with mobile app and desktop versions, were being measured in three separate ways. Product used session counts. Marketing used login events. Finance used billing records. None of them matched, and leadership meetings frequently stalled on which number to trust.
There was no unified activation definition, no funnel visibility, and no way to identify where users were dropping off before they experienced the core value of any product. The team knew adoption was low but had no structured way to diagnose why or measure whether interventions were working.
The question I set out to answer: where exactly are users falling out of the product experience, and what does a fully activated user look like?
The first step was not writing a query. It was getting alignment on definitions. I ran working sessions with Product, Marketing, and Customer Success to agree on a single activation framework, a sequence of behaviors that a user had to complete to be considered genuinely activated rather than simply registered.
With the definition in place I instrumented the full event taxonomy in Mixpanel across all three products, mobile and desktop, and built the ETL pipeline that pulled those events into our data warehouse on a nightly schedule. For the first time, all three products were measured against the same standard.
Once the pipeline was running I ran the funnel analysis across the full user base. The results were stark. Full activation rates were low across the board, and the biggest single drop-off was not at a late step. It was at profile completion, where more than a third of users who had already logged in were stopping.
On mobile the profile completion rate was measurably worse than desktop, which the team had not previously known. The mobile onboarding flow had several more required fields than desktop, a decision made years earlier that nobody had revisited.
Aggregate funnel numbers tell you there is a problem. Segmented analysis tells you where to act. I broke the funnel down by account size and platform to find the highest-leverage intervention points.
Small accounts, which made up 50% of the user base, had the lowest activation rate at 7%. Large accounts, with dedicated staff and more onboarding support, activated at 11%. The platform gap was equally telling: mobile users completed profile setup at a rate 2 points lower than desktop, and the gap compounded through every downstream step.
The funnel showed where users stopped. The next query answered why, by looking at time-to-complete for each step and identifying which steps had the longest median lag, signaling friction rather than disinterest.
The results confirmed what the funnel suggested: the median time from first login to profile completion was 6.2 hours on mobile vs 2.1 hours on desktop. Users were starting the flow, hitting the required fields, and abandoning. Many never came back.
The findings went to Product and the recommendations were direct: simplify the mobile profile setup flow, reduce required fields to match desktop, and add a progress indicator so users could see how close they were to completion. A secondary recommendation was a re-engagement sequence for users who completed first login but had not finished profile setup within 48 hours.
The metrics framework also gave the team something it did not have before: a consistent weekly signal. Every Monday, leadership had a dashboard showing activation rates by product, platform, and segment, with automated alerts when any metric moved more than 5% in either direction.
This was not a dashboard project. It was an alignment project that happened to produce dashboards. The most important work was the conversations that established a shared definition of success before a single query ran. Once the team agreed on what activation meant, everything else followed from that.
The 5x adoption lift is the headline number. The less visible outcome is that for the first time, Product, Marketing, and Finance were looking at the same metrics every week and having the same conversations. That organizational alignment was worth as much as any individual insight the data produced.
Note: All data shown is synthetic and generated for illustrative purposes. The analytical approach, funnel structure, and business outcomes reflect the actual work. Specific figures have been modeled to preserve confidentiality.