This article explores how Trackfly used leading indicators of retention (LIR) to predict customer engagement and inform SaaS onboarding, product development, and go-to-market strategies.
At Stage 2, we spend a lot of time talking about LIR (leading indicators of retention) as a core part of the Catalyst Curriculum for founders. It’s a foundational step in the Science of Scaling framework, and something we ask every company to define before jumping into GTM Fit.
But talking about LIR in theory is very different from implementing it in the real world.
I meet with founders all the time as they work through this process, figuring out what signals actually matter, what’s trackable, and how to build a cadence around reviewing the data. It can be messy, especially in the early days.
That’s why I’m excited to share how Trackfly, a Stage 2 Catalyst company, approached their LIR journey. Brad Duncan, COO at Trackfly, walked through what’s worked, what didn’t, and how they’re using LIR today to guide product decisions, onboarding, and growth.
Before we dive into their story, let’s start with a quick primer on LIR.
What is LIR (Leading Indicator of Retention)?
LIR stands for Leading Indicator of Retention; it’s a customer behavior or signal that happens early in their journey that predicts whether they’ll stick around (i.e., renew or expand). While lagging indicators like churn or retention tell you what has already happened, LIR gives you a chance to intervene early.
In the Science of Scaling framework, we define LIR as:
[Customer Success Leading Indicator] is “True” if P% of customers achieve E event(s) every T days.
Where:
- P = % of customers that meet the behavior threshold
- E = the specific event or action
- T = the time window in which the event must happen
For example, in the early days at HubSpot, one LIR looked like: 80% of customers use 5 features every 60 days. (see an example below)
Why this matters: LIR helps you identify what’s driving retention at the behavioral level, not just financial. It lets you:
- Confirm product-market fit
- Refine your ICP based on who’s succeeding
- Prioritize product improvements or onboarding steps
Early-stage startups (especially those with <15 customers) often can’t do clean cohort analysis. That’s where LIR starts at the individual customer level, then scales to cohorts over time.
For more on Leading Indicators of Retention, you can hop over to this article that breaks it down in more detail.
My conversation with Brad wasn’t just about how Trackfly implemented LIR, it was about the lessons that came with it. Brad shared some of the early missteps, advice for founders getting started, and how LIR is shaping their growth today.
Q: Brad, before we get into LIR, can you give us a quick overview of what Trackfly does and the problem you're solving?
Brad: Sure. Trackfly is focused on solving a big blind spot in the physical retail industry. Most brands and retailers only have visibility into their own performance, they don’t know what’s happening in the broader market. We give them that insight. Our platform shows what's selling, where, and at what rate, so they can make smarter buying and merchandising decisions. It’s all about helping local retailers compete with better data.
As we started to build out our GTM strategy, we needed a way to cut through the noise and pinpoint the behaviors that actually mattered. LIR gave us a way to do that without waiting months to see if a customer churned or not.
Q: I know that a lot of founders struggle to identify what actions to anchor on when it comes to defining their leading indicators. What kinds of actions did you find were strong signals early on?
Brad: Initially, our thinking wasn’t as structured as we would have liked (classic small company stuff). All of us talk to customers, so it was more of a gut feel in the beginning. But over time, we started identifying certain behaviors that felt like strong signals.
One thing we noticed was that if a TrackFly user sends a product invite to connect on TrackFly (e.g., a brand sales rep inviting one of their retailers to connect), that’s not a lightweight action. That user is spending social capital, inviting someone into the experience. Accepting an invite is even stronger. That means they’re in. They’re engaged. So those became important early indicators for us.
We also looked at things like:
- Applying data filters
- Downloading filtered data
Those were real interactions that showed someone was getting value out of the product, not just clicking around and logging in. We wanted to see meaningful actions, actions that suggested they were using Trackfly to make a decision or share insights.
At one point, we also thought that if a customer attended a data review meeting, it was a strong signal (these were regular check-ins where the Trackfly team would walk customers through their own usage data). That felt like a strong engagement moment; they’re taking time to show up, they’re listening. But after a few months, we realized some customers just didn’t need that. They’d already figured the product out and were self-serving (which was even better). So that signal became less important over time.
We learned to ask: Is this action something that actually leads to repeat usage and value? If yes, then we can anchor on that. If not, we move on.
Q: What tools did you use to track your leading indicators of retention (LIRs)?
Brad: We did have to set up some tools in the product to make sure we could track and measure these leading indicators. I had used Pendo before, so we went with that—it’s easy to get up and running. But there are a bunch of other options out there too.
Once we had Pendo instrumented, we combined that product data with the Stage 2 LIR worksheet to build a lightweight but structured tracking system. We created a table where each customer account was listed alongside core usage behaviors. For each account, we measured:
- Whether 80% of users were completing 3 or more LIR actions per week
- Monthly attainment over time
- A color-coded health status (green/yellow/red) to help the CS team quickly see which accounts might need intervention
It takes time to keep it up to date, but for us, it’s worth it. It gives us a quick snapshot of where things stand and lets the team focus their time where it matters most.
Q: Can you share any experiences that validated your LIR approach?
Brad: We had this one retailer who really leaned in. He went in and started sending invites to some of the biggest brands in the industry, all in a row. And in less than 30 minutes, one of those brands signed up. Within the day, we had 6-7 brands sign up because of those invitations.
After that, it became clear that invite activity was one of the strongest signals we had. When someone’s willing to bring a partner onto the platform (especially a big one), it shows they see real value. It’s not just them clicking around; it’s them putting their name on it. That kind of action ties directly into whether they’re going to stick around and keep using the product.
Q: How did LIR impact your customer success strategy and onboarding flow?
Brad: In our customer success motion, it’s about narrowing down and focusing on the high-value items, and getting people to do those sooner. Now that we’ve identified the behaviors that really matter, we can focus on making sure users get there faster.
We can also use those signals to understand “hot” leads during our trial period. That one signal tells us they’re finding value and is a great indicator that we should be reaching out.
Lastly, we also started viewing usage data in cohorts, looking at how long it takes different customer segments to hit key milestones. That helped us understand where people get stuck, what separates high-retention accounts from others, and how to improve time-to-value across different user types.
Q: Tell me about what didn’t work?
In the beginning, we probably tried to track too much. We had all these different actions we thought were important, and we were trying to build this perfect customer profile instead of narrowing in on what really mattered. Looking back, we should’ve started simpler and just focused on the core behaviors that clearly tied to retention.
Q: What advice do you have for founders when it comes to building out and managing your leading indicators of retention (LIRs)?
Brad:
- Talk to your customers.
Talk to your customers, talk to your customers, talk to your customers. It’s easy to sit in the office and think you know what’s going on, but you have to hear it from them. Your sales teams can really help with this, but non-sales roles are critical too because the conversations have a different context without the undertones of a sales motion.” - Focus on repeatable behaviors.
Focus on the things that are actually repeatable and drive value. If it’s not repeatable, it’s not scalable, and that’s not helpful when you’re trying to figure out retention. - Start simple.
Start with just a couple of indicators and be prepared to adapt and adjust. We made the mistake of making it too complicated at the beginning. - Don’t expect perfection.
Even now, we’re still learning. It’s not like we figured out the perfect LIR and everything’s automated. It’s a process, and it keeps evolving. - Look at what signals action.
For us, if they were inviting someone else to the platform, that told us a lot. Find those activities that require someone to invest time, energy or even social capital. Those are the kinds of things worth paying attention to.
Q: What’s next for Trackfly’s use of LIR?
Brad: Right now, we’re trying to improve our visibility into the LIRs for two different customer types that we have, one LIR approach for brands and one for retailers, since they behave so differently on our platform. We’re also working on applying the LIR approach to our freemium/PLG go-to-market strategy to move people through that funnel more effectively.
The best part is, now we’re not guessing. We know what to look for and can actually act on it.