Visibility & reporting
Every report request starts a week of spreadsheet assembly
The Monday ops meeting needs numbers: open orders, late lines, quotes outstanding, margin by customer. Those numbers get assembled Sunday night—pulled from the ERP, patched with spreadsheet data, reconciled by hand, and pasted into a deck. By the time the meeting happens, the data is three days stale and nobody fully trusts it.
When a new question comes up—'which customers are slipping?'—the answer isn't a report. It's a project.
What it looks like
- Recurring reports are hand-built every cycle: export, paste, fix the formulas, reconcile the mismatches, repeat next week.
- Three people bring three different versions of 'sales this month' to the same meeting, and twenty minutes go to whose number is right.
- Every new leadership question triggers days of analyst work, because the data exists but was never organized to answer it.
- Reports describe last month. By the time a trend is visible in the deck, the operation has been living it for weeks.
- The person who builds the reports is also the only one who knows which adjustments the numbers need—and they're on vacation the week the board asks.
Why it happens
The data for any real operational question lives in several places: the ERP for orders and invoices, spreadsheets for pricing and commitments, email for exceptions, niche systems for shipping or production. No single system holds the answer, so every report is a manual join across sources.
The second problem is definitions. 'Late orders' means one thing to sales and another to the warehouse; 'margin' depends on which costs someone remembered to include. Without agreed definitions encoded somewhere, every report author makes their own calls—and the meetings litigate definitions instead of decisions. Reports stay hand-made because the infrastructure for making them otherwise was never built.
What it costs
The honest answer is that it depends on your volumes, your margins, and how long it has been going on. What is consistent across companies is where the cost shows up:
Labor
Analysts and managers spend days each month assembling recurring reports—your most expensive people doing data plumbing.
Stale decisions
Decisions run on last month's picture. Problems that a current view would catch early get discovered at the monthly review.
Meeting waste
Time meant for decisions goes to reconciling competing numbers and re-litigating definitions.
Late signal
Customer slippage, margin drift, and backlog changes surface weeks after they start—when options are fewer and more expensive.
How a purpose-built system fixes it
A purpose-built operational dashboard starts with definitions, not charts. The handful of numbers the business actually runs on—open exceptions, late lines, quote pipeline, margin by customer—get agreed once, encoded, and computed the same way every time. The arguments about whose number is right end because there's one number and everyone knows what it means.
Then the plumbing: the dashboard pulls from the ERP and the other source systems on a schedule, so the Monday meeting looks at this morning's picture. The aim isn't more reports—it's that recurring questions stop being projects, and exceptions surface while they're still cheap to fix.
Concept Interface
Ops briefing — this morning
The recurring questions, answered once, current daily.
| Measure | Today | Change | Note | Status |
|---|---|---|---|---|
| Late order lines | 17 | +4 vs. last week | Two vendors driving it | Watch |
| Quotes awaiting approval | 9 | −2 | Oldest: 3 days | Aging |
| Open customer claims | 5 | No change | One awaiting credit memo | Customer update due |
| Backorders past promise | 2 | −3 | Customers notified | Handled |
Example System — one agreed definition per measure, pulled from source systems, not assembled by hand.
The analysts don't disappear either. Freed from assembling the recurring numbers, they answer the genuinely new questions—the ones worth a person's week.
When a simpler fix is enough
If your truth lives mostly in one system, start there: the ERP's own reports or a general BI tool pointed at clean data may cover most of it, and both are cheaper than custom. A shared definitions page—what 'late' means, what 'margin' includes—fixes half the meeting problem for free.
Custom earns its place when the real answers span systems—ERP plus spreadsheets plus shipping plus email-based exceptions—and when leadership wants exceptions surfaced daily, not assembled monthly. That is the gap an operational dashboard built on your actual systems closes.
Related systems and industries
Common questions
- For data that lives in one clean source, you should—a BI tool on top of a healthy ERP is the right answer and cheaper than custom. BI struggles where most operational truth does: in spreadsheets, inboxes, and niche systems, with definitions that were never written down. The tool isn't usually the hard part; the sources and the definitions are. When those are the gap, the work is building the operational layer underneath—and then almost any front end will do.
- By deciding once, in writing, with the people who currently argue. 'Late' gets a rule. 'Margin' gets a formula. The dashboard encodes the decision and shows it—every measure carries its definition—so disagreement shifts from 'whose number is right' to 'is this the right rule,' which is a useful argument that only needs to happen once.
Your operation changes daily. Your reports describe last month.
A Workflow Review maps where your recurring numbers actually come from—and what current, agreed versions would take.