Academy · Data quality · Method
A number is only as good as its data,
and you can check that in ten minutes.
Before a number goes into the monthly review, a short check pays off: is anything missing, is anything duplicated, do units and currencies match, does the data cover the period you're talking about? This method sets out four checks you can run on any spreadsheet and shows them with fictional examples.
Which question must the source answer?
For each important field, count how many rows are empty. Keep empty (not filled in), null (deliberately no value) and missing (the field doesn't exist on that row) apart. Peak Analytics keeps the three separate; in your own spreadsheet you have to do it yourself.
A number is only as good as its data,
and you can check that in ten minutes.
Complete, unique, consistent, current.
Before a number goes into the monthly review, a short check pays off: is anything missing, is anything duplicated, do units and currencies match, does the data cover the period you're talking about? This method sets out four checks you can run on any spreadsheet and shows them with fictional examples. Four questions for any data basis: are values missing? Are there duplicates? Are units and currencies the same? Does the data cover the period you're discussing?
Finding & data slice
What is proven so far.
No finding. This guide shows a way to check, using fictional examples; it rests on no dataset and no study.
ObservationInterpretation
决策
The evidence is still pending
No measurements. The figure stays empty until real, checked data exists.
Transparency is part of the result
Count, reconcile, label.
Three steps that turn "this number feels wrong" into a named cause.
Count the gaps
For each important field, count how many rows are empty. Keep empty (not filled in), null (deliberately no value) and missing (the field doesn't exist on that row) apart. Peak Analytics keeps the three separate; in your own spreadsheet you have to do it yourself.
- Reconcile with a second source
- Compare a total with an independent place: orders with incoming payments, receipts with the bank statement, tasks with the board. If they differ, look for duplicates, cancellations and different date fields.
- Label units and data date
- Check currency, unit and time zone. Mixed currencies are not added up but shown separately. Write the data date on the number so nobody compares half a month with a whole one.
What counts as an
observation?
share of empty required fields · counterparty missing on 7 of 120 receipts
Event defined reference groupMeasurement scheme · no calculated study value03 / Consistentcurrencies and units · euros and francs kept apart
How data becomes orientation
Observation. Proof. Next step.
Count the gaps
For each important field, count how many rows are empty. Keep empty (not filled in), null (deliberately no value) and missing (the field doesn't exist on that row) apart. Peak Analytics keeps the three separate; in your own spreadsheet you have to do it yourself.
Key idea7 receipts completed, the totals match
Reconcile with a second source
Compare a total with an independent place: orders with incoming payments, receipts with the bank statement, tasks with the board. If they differ, look for duplicates, cancellations and different date fields.
Key ideaReported separately per currency
Label units and data date
Check currency, unit and time zone. Mixed currencies are not added up but shown separately. Write the data date on the number so nobody compares half a month with a whole one.
Key ideaDifference explained: Friday's data versus Monday's
Your quality check
Would your number survive a follow-up question?
Tick what you have checked. Your selection stays in this view only.
0 of 3 points checked
What this article is based on.
Editorial page with fictional examples (Willowcourt Advisory (fictional), Nordwind Travel Goods (fictional), Harbourview Agency (fictional)). The guide describes an approach; where Peak Analytics covers a step today, the page says so explicitly. No measurements, no customers, nothing is sent or saved.
- Editorial demo
Synthetic numbers in the browser. No real data is read, computed or saved.
- Checked evidence
Peak Analytics is a preview: pivot over orders, receipts, tasks, goals and files; count, sum, average, minimum, maximum; day to year; CSV export; sharing a released snapshot; largest value as a task.
peak-one app/w/[scope]/apps/analytics, lib/bi/model.ts, lib/views/pivot.ts, lib/bi/share.ts · checked 2026-10-29 - Evidence pending
No dataset and no study. The page shows a way to check, not measured values.
Check first, then report.
A number with its data date and known gaps is more honest than a smooth number with neither.
Frequently asked questions
Do I need Peak Analytics for this guide?
No. The guide works with any spreadsheet. Peak Analytics is a preview; where it covers a step today, the page says so.
What is the difference between empty, null and missing?
Empty means not filled in, null means deliberately no value, missing means the field doesn't exist on that row. Treating them alike distorts averages.
Does the check prove the data is right?
No. It finds the usual errors. Values that are complete but wrongly entered only show up when you reconcile with a second source.
Does Peak Analytics check data automatically?
Partly: the preview never adds mixed currencies, keeps empty, null and missing apart and marks cut-outs. Anomaly detection and alerts are planned.