The fastest way to lose confidence in a CRM is two dashboards that disagree. We fix the definitions first, then build the reports, then retire everything that contradicts them.
In short
Dashboards disagree because the same metric has been implemented differently in each one: different date properties, different filters, different lifecycle assumptions. The fix is a written metric dictionary with one owner per number, implemented as shared reports, with the contradictory legacy dashboards retired rather than left running alongside.
Signals
Someone exports to a spreadsheet every month because the dashboards cannot be trusted as-is.
Forty dashboards exist, four are used, and the rest still show up in search results.
You can see how many deals closed but not where in the funnel they were lost.
Marketing and sales credit the same revenue twice with no agreed model.
What you get
Definitions, calculation logic and an owner for every number that appears in a leadership meeting.
ExploreThree layers: an executive view, manager views for coaching, and rep views for daily work.
ExploreConversion by stage, time in stage, and cohort views so you can see where deals actually die.
ExploreAn attribution model everyone has agreed to, and forecast views based on stage criteria rather than optimism.
ExploreHow it runs
Every existing dashboard catalogued, and every leader asked which number they actually use to make decisions.
Working sessions until each contested metric has one definition and one owner, written down.
New reports built on the agreed definitions, and the old conflicting ones deleted. Leaving them running defeats the exercise.
Quarterly review of usage and definitions, plus data quality monitoring so reports fail loudly rather than quietly.
Questions
The questions we get asked on almost every call about this work.
For most mid-market revenue reporting, yes. You need a warehouse and a BI layer once you are joining HubSpot data with product or finance data at scale — we will tell you honestly which side of that line you are on.
Almost always a different date property, a different filter on deal type, or one view including a pipeline the other excludes. The metric dictionary exists to make that impossible.
One named person per metric, usually in RevOps or finance. Shared ownership means no ownership, which is how definitions drift.
Only with your agreement, and only after the replacements are live. But we do push for retirement — parallel contradictory reports are the root cause, not a safety net.
Two to four weeks typically. The building is fast; agreeing the definitions is what takes the time.
Next step
Send us the two dashboards that disagree. Explaining the gap is usually the first ten minutes of the engagement.