Defining KPIs At An Early-Stage Startup
A Series B B2B SaaS company reached a point where they were collecting plenty of data but could not make sense of what any of it meant for the business. They had event tracking in place and a long list of surface-level KPIs, but none of it helped them understand whether the product was improving in ways that would actually drive company growth. Most reporting was manual. The company had no data warehouse, no clear metric definitions, and no alignment on what success looked like.
They had data, but they weren’t using it to build a story and hone in on what mattered most.
The Problem
Before I stepped in, the team relied on very basic metrics that came directly out of their analytics tool. These were updated by hand and were not tied back to revenue in a meaningful way. The product team was shipping improvements without knowing how to evaluate the impact. Leadership wanted to understand whether changes in the product translated into real progress, but the existing KPI set did not tell that story.
The biggest issue was simple. They were optimizing for user counts, yet a large share of users came from regions that rarely retained or converted. High growth on paper did not translate into revenue growth for the business.
My Approach
During the first month, I focused on building the foundation that would allow us to answer the bigger questions. This meant standing up a proper data warehouse, pulling key data streams into it, and moving the company away from manual metrics updates. Once the basics were in place, I spent time with leadership and the product team to understand the customer journey and the real drivers of revenue.
It became clear that the company needed a metric framework built around the customer lifecycle, not raw user growth. The goal was to know which users were actually valuable and what behaviors predicted long-term retention and paid conversion.
The Solution
We rebuilt the KPI structure around three simple stages.
Activation
We identified usage thresholds that predicted long-term engagement. Any new user who reached that threshold was considered activated.
Retention
Among users who activated, we measured how many continued to engage at a level that demonstrated real value.
Conversion
Among retained users, we measured how many became paying customers.
These three metrics gave the company a clean and reliable way to understand progress. I documented every definition, translated them into the warehouse and the modeling layer, built dashboards around them, and automated the reporting so the team no longer depended on manual updates.
When leadership saw the new KPI framework, the reaction was immediate. They realized they had been tracking many things, but not the things that actually moved the business forward.
The Outcome
Once the company aligned around activation, retention, and conversion, everything else became clearer. Product teams started focusing on the behaviors that predicted long-term value. Growth efforts shifted toward acquiring users who were more likely to convert. The revenue model itself was adjusted so it better reflected the way valuable users moved through the funnel.
Most importantly, the team could finally connect product decisions to revenue outcomes. They understood which improvements mattered and which were distractions.
Over the following year, the company saw its ARR increase roughly 4x. Leadership attributed much of this improvement to having a clear and consistent view of what success looked like and how to measure it.
