“We didn’t sign up to be a dashboard vending machine. We signed up to create impact.”

That one sentence sums up the frustration many data professionals feel today. You work hard, build dashboards, share metrics but still find yourself sidelined when big decisions are made. You’re stuck in the Dashboard Factory. Endless requests, shallow impact, little recognition.

The good news? There’s a way out. I’ve lived it.

Today, I’m sharing the 3 most important moves that helped me break free from the Dashboard Factory and start driving real, visible business impact.

Why This Matters

If you're a data analyst, scientist, engineer, or leading a data team in a startup or scale-up, chances are you've felt it: the low-leverage trap of just building dashboards.

But the data profession is meant to do more:

• Influence decisions.
• Drive measurable results.
• Gain visibility with senior leaders.

This newsletter is your blueprint to move from reactive service provider to strategic business partner.

How I Realized I Was Stuck

In my previous newsletter, I described the reality many data leaders face: being reduced to nothing more than a dashboard producer. I was there too. Building what people asked for, but rarely what they needed. Working hard but always being downstream of decisions.

It was frustrating. And pointless.

So I made a decision: I was going to find a way out. I didn’t want to just answer “what happened.” I wanted to help answer “why” something happened, and ultimately, shape what should happen next.

That journey led me to three key changes I’ve ever made in how I work with data.

The Three Moves That Changed Everything

1. Self-Service That Actually Works

Yes, self-service. The term has been overused - and often misunderstood. But when done right, it’s not a buzzword. It's not a myth. Not a scam. It’s a game changer.

Here’s my take:

• Dashboards should only answer what happened, and there should be very few of them.
• The why did it happen should be explored directly by trained business decision makers - what I call super users. (Marketing Managers, Product Managers etc)
• The goal: in each domain, super users should be able to answer 80% of their “why” questions without involving the data team.

How I make this work:​
→ Give business users direct access to a well modeled layer in the Data Warehouse (I use denormalized, one-big-table interfaces)
→ Let them use the tools they already love (in my case mostly Google Sheets).
→ Leverage integrations like Google Sheets + BigQuery to enable direct querying.
→ Train them not to just read dashboards, but to explore hypotheses.
→ Create sandbox environments where they can experiment safely without creating a wild growth of unverified reports and analyses.

→ Only if they want to publish something company-wide, they need to go through the data team

This is about shifting the mindset from data delivery to data empowerment.

And this shift alone - giving smart, motivated business people the tools and space to explore - freed up my team from constant ad hoc requests and brought us closer to the business.

2. Own Experimentation End-to-End

The second big unlock: take full ownership of experimentation.

Why? Because experiments have very tangible, easily measurable business impact. Unlike reporting, which often gets ignored or debated, experiments show clearly whether your work drove results.

As my friend and Senior Director Analytics & Data Science at Doordash, Kshira Saagar, puts it:

“You don’t exist here to just measure something. You exist here to see if something is truly incremental.”

Why this matters:

• Experiments prove the value of data work.
• They give you a seat at the table by default.
• They build credibility - fast.

Real-world example:​
At Pizza Hut Digital Ventures, Tris Burns built experimentation into the data team’s core offering. They started from scratch and ended up running 1–2 new experiments per week. One standout test: removing their forced pizza customization flow and adding a simple “Add to Basket” button. The result? An ugly button, yes - but it boosted revenue by over £1M. Tangible. Undeniable.

My own experience:​
At Rocket Internet, the world's largest internet venture builder, we tested recommendation engines for Lazada (the Amazon.com of South East Asia, exited to Alibaba for $2bn). Off-the-shelf tools failed to generate incremental gains. So we built a semi-manual recommender. Not fancy, but it outperformed everything else - and saved us a ton in tooling costs. Business impact squared.

Takeaway:​
If you want to show that data matters, own experimentation. Help design, execute, and interpret tests. Push for iteration. Bring the numbers and the insights.

3. Send Data to Systems, Not Dashboards

Stop treating dashboards as the final destination.

Dashboards show numbers. But business leaders want to use numbers to make decisions - often automatically.

That’s where the real leverage is.

What this can look like in practice:​
At Zalando, Europe’s leading online fashion platforms, the turning point was when they started optimizing ad spend not for revenue - but for projected lifetime value (CLV) based on a deep understanding of the full user journey. While competitors optimized on shallow metrics (like revenue before returns), Zalando focused on profitability. That edge helped them dominate the market. And it was driven by the data team that:

• built a system that understands the impact of each touchpoint in a user journey preceding a purchase
• weighting the impact of each touchpoint based on its significance (something that is common today but was unheard of in 2009)
• sent the CLV data to Google Ads and Meta Ads.

Why this matters for data teams:

• It’s high-leverage work that scales.
• It directly impacts profitability, CAC, CLV, retention, and more.
• It transforms data from a passive reporting layer to an active decision layer.

If you’re serious about moving past the Dashboard Factory, this is the frontier.

Bottom Line

Escaping the Dashboard Factory isn’t about working harder. It’s about working differently.

These three shifts helped me break free and start creating real business impact:

1. Self-service that empowers super users - so the data team can focus on higher-leverage work.
2. Owning experimentation - to prove incremental value and drive strategy.
3. Sending data to systems - to make automated decisions, not just pretty charts.

Each of these requires trust, technical setup, and a mindset shift. But they’re all doable and incredibly rewarding.

So ask yourself:

Are you building dashboards… or driving decisions?

Choose wisely.

See you next week!

Sebastian