Every Monday it starts again. Someone exports from the CRM, someone else pulls the ops sheet, a third person reconciles the numbers in a master spreadsheet that only they understand. By Thursday the data is already stale. By Friday you are presenting numbers you do not fully trust to a room that has stopped asking hard questions. We have seen this pattern at fourteen of the last twenty clients we onboarded. The teams were not lazy, they were stuck. The loop had become invisible, part of the job. Nobody had time to fix it because fixing it required the same three hours the loop was already eating. This is what we found when we looked at the data source chain, and what it cost to untangle it.
The loop starts on Monday. Someone opens the CRM, exports last week’s activity, drops it into a shared folder. Someone else opens the ops spreadsheet, the one that lives on the shared drive and has not been meaningfully updated since 2022, and copies the relevant rows into the master. A third person, the one who actually understands the master, runs the formulas and produces the number. By Tuesday morning, that number is in the weekly deck.
By Thursday, it is stale.
The data it is built on was already three to five days old when the export ran. The CRM only reflects deals someone remembered to update. The ops sheet reflects whatever the last person to open it had time to fill in. The master formula has an error in column K that everyone knows about and nobody has fixed because fixing it would mean understanding column K, and understanding column K would take a morning no one has.
On Friday the deck goes to leadership. The numbers are presented with confidence. The questions have stopped being hard.
Why the loop persists
We have seen this pattern at fourteen of the twenty clients we onboarded in the last year. The teams were not disorganised. Several of them had strong data people. The issue was not capability, it was structure.
The reporting loop persists because breaking it costs exactly the time it is already eating. To fix the Monday export, you need to spend a morning mapping the data model. To fix the master spreadsheet, you need someone who understands column K to sit down for a full day and document what it is actually doing. To move off manual reconciliation, you need to agree on a single source of truth for four different things (deal status, pipeline value, ops throughput, resource utilisation) that four different systems currently track in four different ways.
None of that can happen in the thirty minutes between the Monday export and the Tuesday standup. So the loop runs again.
What the data source chain actually looked like
In one typical case, a forty-person professional services firm, the weekly report touched six separate data sources before it reached the deck.
The CRM held deal stage and forecast. The project management tool held current utilisation and hours logged. The accounting system held actuals. A shared Google Sheet held the ops team’s capacity plan. A second Google Sheet, the one built by someone who had since left the company, held the reconciliation logic that tied all of it together. And there was a Notion page, not a data source exactly, but treated like one, where the ops lead kept manual notes on which numbers in the sheet were trustworthy and which were not.
The weekly report required touching all six. In the right order. By the right person. Without anyone else saving a file at the same time.
The total time: about three hours, spread across three people, every week. The cost, at fully loaded day rates: roughly €800 per month in labour producing a report that everyone presenting it quietly knew was approximate.
If that number sounds familiar, you can run the same calculation for your own team. The ROI calculator takes three inputs, hours per week, headcount involved, and average day rate, and shows the annual cost, the datareaches alternative, and the payback period.
What it cost to untangle it
The fix took seven days. Not because the data work was simple, it wasn’t, but because the scope was tight. We built one dashboard that connected directly to the CRM, the project management tool, and the accounting system. The reconciliation logic from the old spreadsheet was rebuilt as a proper data model: transparent, documented, testable. The capacity plan from the shared Sheet was migrated into the same structure.
We did not try to fix the Notion page. We asked the ops lead to tell us, explicitly, which numbers she trusted and which she adjusted manually before using. Those adjustments became named rules in the data model. The ones that could not be formalised became comments. Nothing was hidden.
The deliverable on day seven was a single screen. Every number the weekly deck had required was on it, updated automatically when source data changed, with the reconciliation logic visible to anyone who wanted to understand it.
What the reporting rhythm looks like now
The Monday export still happens, some systems do not push data, they have to be pulled. But it takes four minutes instead of forty. The person doing it is not the one who understands the master spreadsheet; it is whoever is in the office first.
The Tuesday deck is produced from the dashboard. The numbers are the same ones leadership will see if they log in themselves, which they occasionally do now that the dashboard exists. The questions have started being hard again, which is uncomfortable in the short term and useful in every other way.
Column K is gone. Nobody misses it.
The ops lead told us, three weeks after go-live, that the strangest part was not the time saved. It was the absence of the low-grade anxiety she had carried into every Friday presentation for two years. The anxiety of knowing that if someone asked the right follow-up question, she would not have a good answer. That feeling had become so normal she had stopped noticing it until it was gone.