Qasa isn’t a media company. We’re a relatively young, product-driven company with a small marketing team and an even smaller data team. Historically, the two teams haven’t really orbited the same planet.

Marketing needs fast, compelling stories. Data wants rigorous analysis, usually in service of the product. Two goals that tend to collide more than they align. The relationship between data and marketing was always polite, but distant.

Marketing wanted visibility. They wanted to get our data into the press, to ride the news cycle with timely, relevant insights. The problem? The data team was busy serving product needs; modeling, testing product changes, and debugging pipelines. Media angles weren’t built into our workflow.

Worse, the times marketing did try to collaborate with data, it turned into a fire drill. Or a stale annual report. Or a Slack message left on read. 

The breakthrough didn’t start with a query or dashboard. It started with a conversation. We sat down with our Communications Director and realized something: it wasn’t that they didn’t care about data quality – it was that they didn’t have access to the right data, at the right time, in a way that mapped to how journalists think.

And we weren’t helping. We were guarding the data, insisting that every number be validated, hiding metrics behind complicated SQL, and ignoring the fact that the underlying data was constantly changing as the product evolved.

To make the data reliable – even as the product evolved – we built materialized views, which are basically curated, stable datasets that insulated marketing (or any consumer) from the chaos of raw production tables. They acted as a contract between teams, ensuring that a chart wouldn’t break just because a developer renamed a column or changed a home status enum. With that foundation, we built a set of self-serve dashboards tailored to marketing's needs, e.g.:

We focused on metrics that were general enough to tell most stories the market threw at us. Not “rental price delta for 2-room apartments in Uppsala with balconies,” but rather general market indicators that could support a wide range of headlines. Things like:

We didn’t just ship dashboards. We built confidence in the data, in the definitions, in marketing’s ability to use them, and in the data team’s trust in the marketing team.

The moment it all came together, when we realized just how far we’ve come, was when Northvolt’s bankruptcy made national news. Our head of Comms spotted the angle immediately. She pitched the story to a journalist at Dagens Nyheter, Sweden’s biggest paper, provided data directly from our dashboards, and was interviewed as an expert source. The story ran the next day – with Qasa at the center. That’s when we knew it was working. Data wasn’t a blocker anymore. It was fuel.

The collaboration transformed both teams:

  • Marketing got faster, more confident, and better at telling data-driven stories.
  • Data got fewer fire drills, less anxiety about data quality, and a stronger sense of impact. Even morale went up. It’s more fun to build dashboards when they’re landing headlines.
  • Qasa got more press, better alignment, and a shared understanding of how our platform fits into the broader housing market. It also helped Qasa reinforce its role as a thought leader. We’ve become more active in the public housing debate – providing unique insights to journalists, politicians, and policy makers alike. We’re not just reacting to the news cycle anymore; we’re also shaping it. Our Head of Communications was even named one of Sweden’s top communicators of 2024, a sign that the stories we’re telling are resonating, and that Qasa is seen as a credible, data-backed voice in the national housing conversation.

If you’re a data team at a product-first company, and you want to support marketing without drowning in ad hoc requests, don’t just build what they ask for. Sit down and listen.

Context matters more than the query. Design for generality. Metrics should flex with the news cycle. Decouple your data pipelines from product chaos. Invest in an analytics layer (materialized views) or similar to give marketing a stable playground. Celebrate the wins. A story in the paper beats a spike in dashboard views.

It started with answering a Slack message — but it only worked because we built something that non–data scientists could actually understand, use, and communicate to the world. No, we’re not a newsroom. But our data lives in the real world. It tells stories. And now, we have a team that knows how to tell them.