SELECT warm_feelings, kind_words FROM case_studies WHERE customer='JustPark'

How JustPark Rebuilt Their Analytics Stack Around Count. And Left Looker Behind.

Being a data leader in a company mid-merger is hard enough. Now add a looming BI renewal, two conflicting tech stacks, and the realization that your primary tool no longer fits your team.

That was the situation Padraig Alton, Data Lead at JustPark, found himself in.

“After years of relying on Looker, we hit a breaking point—too expensive, too rigid, and no longer fit for our evolving analyst-led culture.”

The merger brought urgency. The costs pushed things over the edge. And what followed was a transformation—not just in tooling, but in how the data team worked.

Here’s how JustPark made Count their new home for analysis, modeling, and more.

From centralized self-service to embedded intelligence

JustPark’s data team used to operate a self-serve model, centered around Looker. It worked—for a while.

But as the business matured, and the merger brought in new ways of working, they shifted to an embedded analyst model. Each team—marketing, ops, product—had their own data specialist, with the central data team focused on modeling, infrastructure, and support.

“Domain knowledge is so important… it takes a really long time to take a single analyst and get them to understand the context for every single team in the organization.”

This model worked better for JustPark’s scale and speed. But it required different tooling—especially something that could give analysts freedom without sacrificing structure.

Why Looker had to go

The decision to migrate off Looker wasn’t just about preference. It was practical.

“It was hard to make the case that [Looker] was really good value for money when there are a lot of new entrants into the BI space and many of them are offering much lower annual fees. At some point you kind of find it hard to justify to yourself.”

With Looker renewal costs rising, and Tableau already in use at the parent company, the team had a decision to make.

Plus, LookML (the language used to model data in Looker) created additional friction. Migrating off Looker meant rebuilding the entire modeling layer.

Count gave them a more flexible way forward. With templated SQL and dbt, they could untangle the old models and start fresh—with transparency.

Modeling a way out of Looker

The hardest part of the Looker migration wasn’t dashboards, it was the modeling logic locked inside LookML.

Padraig and his team needed to rebuild it all, from the ground up, using SQL and dbt. An Analytics Engineer was dedicated to this for nearly six months, carefully translating and restructuring what had been tightly bound inside Looker.

“It took months of one person just sitting there trying to figure out how to remodel the data out of Looker properly.”

Count made that process visible and manageable. It became the thinking space: a place to pull apart Looker’s generated SQL, validate assumptions, and rework logic as modular building blocks.

“Count is quite a multi-tool… like a Swiss army knife. It’s been great for data modeling during the migration but also in so many other areas for exploration.”

The visual structure of the canvas let them trace relationships, model transformations, and bring clarity to what could’ve been chaos.

Count as the analyst’s power tool

With the modeling layer rebuilt, analysts at JustPark leaned into Count for far more than transformation logic.

They began using it to:

  • Model complex queries
  • Visualize dependencies across logic layers
  • Build out freeform explorations and internal narratives

The canvas format encouraged this kind of work. It supported visual thinking, async collaboration, and quick context-switching between analysis and communication.

“Our analysts are using it in completely different ways… You can see almost these chaotic mind maps of Count cells and sticky notes and data moving between different bits and into visualizations as people try and figure out solutions to complex problems.”

In essence, Count gave JustPark analysts a space to think out loud—with SQL.

Enabling new kinds of thinking

Beyond migration, Count helped JustPark rethink how they analyze and communicate data.

Padraig shared an example inspired by Duolingo’s lifecycle modeling. His team built a custom lifecycle map for subscription customers, visualizing transition states and revenue implications. It helped them:

  • Track churn and reactivation behavior
  • Understand revenue impact by user segment
  • Layer in seasonality and contextual factors
“You can really get a lot of insight out of a view like this… I chucked this into Count more as an experiment, but I’ve been really pleased with what we were able to put together.”

This type of analysis would’ve been nearly impossible to build—and explain—in traditional dashboard tools.

Lessons from the migration

There’s no way around it: migrating from Looker is a big lift.

One of the hardest parts wasn’t just rebuilding dashboards. It was moving the logic out of LookML and into dbt—while keeping it understandable.

“Looker really leans on LookML… so just having the ability to pull out the SQL and think about how to remodel this in a better way—I found that invaluable.”

Count played a critical role here. It let the team:

  • Deconstruct Looker logic
  • Write and test dbt models visually
  • Validate transformations step-by-step

Final thoughts

JustPark didn’t just switch tools. They changed how their analysts work.

Count gave them a way to:

  • Collaborate across teams
  • Explore ideas visually
  • Build modeling workflows outside of rigid BI interfaces

Tableau still plays a role for some stakeholder-facing dashboards. But Count is where the work happens.

If your data team is feeling boxed in by traditional BI—especially if Looker isn’t pulling its weight—Count might be the rethink you need.

“It’s the flexibility of Count that made the difference for us. It met us where we were and helped us get to where we wanted to go.”