Everyone wants a dashboard. Nobody wants the same one.

Ask five people for “the dashboard” and you get five different requests wearing the same word. Marketing wants lead source. Sales wants cycle time. Finance wants cost per opportunity. Success wants retention. Same CRM underneath, four conclusions, and a meeting that solves nothing.

That isn’t a data problem. It’s a translation problem — and it’s the last mile almost nobody builds.

The same row means different things to different people

Take one deal. Marketing sees the webinar that sourced it and counts a lead at €40. Sales sees ninety days of work and a cycle two weeks longer than the quarter allows. Finance sees a cost per opportunity that only makes sense if it closes. Success sees an account that will churn in eight months if onboarding slips.

Nobody is wrong. They’re reading different columns of the same row and each calling it “the truth”. Lead source, conversion touchpoint, time to convert, cost per lead, cost per opportunity, average customer lifecycle — those aren’t competing metrics. They’re different questions, and each team only has the budget to act on one of them.

Clean data isn’t the deliverable. A decision is.

Most data projects stop one step early. The CRM gets cleaned, the fields get governed, the dedup runs — and then everyone is handed the same report and told to be data-driven.

Clean data is the raw material, not the product. The product is a view a specific person can look at on a Monday and change what they do. If a dashboard doesn’t end in a choice, it’s decoration with a refresh button.

Everyone sees the same thing. Each person sees their part.

Democratising data doesn’t mean giving everyone access to everything. That’s how you get four people building four spreadsheets from the same export and disagreeing by Thursday. It means each team gets a view built for the decision they own — and every view reads from one governed source.

Marketing’s view answers: which sources produce customers, not leads? Sales’: where do deals stall, and how long does “normal” take? Finance’: what does an opportunity cost, and when is it worth it? Success’: which accounts are drifting, and how early can we tell? Different numbers, same row. That only holds if the stages and definitions mean the same thing underneath — otherwise you’ve democratised a disagreement.

The company only has three numbers

Under all of it sit three questions, and they’re the ones the board actually asks.

What does it cost to win a customer? Not cost per lead — cost per customer, including every deal you lost getting there. What does it cost to keep one? Support, success, the discount you quietly gave at renewal. What does it cost to lose one? The one nobody calculates: the revenue, the cost of replacing them, and the referrals that never happened.

Every team metric should ladder into one of those three. If cost per lead is falling while cost per customer climbs, marketing is winning its scoreboard and losing the company’s. That’s not a reporting quirk — it’s the entire reason the views have to sit on the same data.

How you actually build it

Start from the decision, not the chart. Ask each team what they’d do differently if a number moved. If the answer is “nothing”, don’t build it. Then agree the definitions out loud, once: what counts as a source, when the clock starts on time-to-convert, what makes an opportunity an opportunity. Write it down. That document is worth more than the dashboard it produces.

Then give each team the smallest view that supports their choice, and one shared view showing win, keep, lose. All of it reading from the same CRM people actually use, so when two teams disagree, the argument is about the decision — not about whose export is right.

Data democratisation isn’t everyone seeing everything. It’s everyone seeing the same thing, and each person seeing the part that’s theirs to act on.