Sign in

Blog · Board and management reporting

Dashboard standards: definitions and data checks

Set an internal acceptance standard for dashboard definitions, period comparisons and source reconciliation. Use a small revenue example and spreadsheet formulas to check whether a dashboard is ready for business decisions.

The short answerDashboard standards are an internal acceptance agreement for what each measure means, which periods and records it covers and how its totals reconcile to approved sources. Document the definitions, calculate a worked control example and record failed checks before release. Give each measure an owner and make any limitations visible beside the result.

Dashboard standards define when business figures are ready to use: the measures are documented, periods are comparable and totals reconcile to approved sources. Treat the standard as a repeatable acceptance check with an accountable business owner. A chart's appearance cannot establish whether its numbers mean what the reader thinks they mean.

Start with a measure acceptance card

For every decision-driving measure, write down the following before reviewing the chart. The example is a US distributor's monthly net revenue measure.

Field Acceptance requirement Example
Business meaning State what the measure answers Revenue posted in the reporting month
Calculation Name the columns and operations Sum signed net-revenue amounts
Scope Identify included and excluded records US entity, USD, revenue accounts; no sales tax
Date basis Choose the date used for selection Ledger posting date
Period Set complete, comparable boundaries September 2026 versus August 2026
Source Identify the approved export Closed revenue ledger summary
Owner Identify the person who approves meaning Controller
Control Define evidence for acceptance Source total and dashboard total agree

Specify treatment of returns, credit memos and late entries. A signed net-revenue column is useful only if the source owner confirms what it contains. Keep definition changes dated; the existing guide to versioned metric definitions covers that governance.

The GSA's performance-data verification directive ties verification and validation to confidence that information is credible for its intended use. That directive applies to GSA. The acceptance process here is an internal business standard your organization can adapt.

Recalculate from the approved export

Place the following ledger summary in a worksheet named Source, with the header in row 1. Each identifier represents a distinct exported summary record; amounts are already net of applicable credits. These are all the records in the example.

Record ID (A) Posting date (B) Net revenue (USD) (C)
AUG-A August 31, 2026 100,000
AUG-B August 31, 2026 80,000
AUG-C August 31, 2026 70,000
SEP-A September 30, 2026 120,000
SEP-B September 30, 2026 90,000
SEP-C September 30, 2026 90,000

Store column B as actual spreadsheet dates. In a separate control worksheet, calculate September revenue in B2:

=SUMIFS(Source!C2:C7,Source!B2:B7,">="&DATE(2026,9,1),Source!B2:B7,"<"&DATE(2026,10,1))

Calculate August revenue in B3:

=SUMIFS(Source!C2:C7,Source!B2:B7,">="&DATE(2026,8,1),Source!B2:B7,"<"&DATE(2026,9,1))

The upper boundary excludes the following month. This approach also works if posting values include a time of day. Microsoft's SUMIFS documentation confirms that the function sums values meeting multiple criteria.

Check the figure and the comparison

The September control is 120,000 + 90,000 + 90,000 = $300,000. August is 100,000 + 80,000 + 70,000 = $250,000. Compare complete months using the same date basis and inclusion rules.

Calculate growth in B4 and format it as a percentage:

=IF(B3=0,"Not comparable",(B2-B3)/B3)

Growth is (300,000 − 250,000) / 250,000 = 20.0%, rounded to one decimal place. A zero prior value returns a label because percentage growth would not be defined.

Now suppose a proposed dashboard shows September revenue of $288,000. Enter that displayed amount in B5 and calculate its reconciliation difference in B6:

=B5-B2
Acceptance check Approved control Proposed display Result
September net revenue (USD) 300,000 288,000 Difference of −12,000; investigate
August net revenue (USD) 250,000 250,000 Agrees
Month-over-month growth 20.0% 15.2% Uses the unreconciled September figure

The proposed growth is (288,000 − 250,000) / 250,000 = 15.2%. Its arithmetic is consistent with the display, but the display is inconsistent with the approved source. Reconcile the $12,000 difference before accepting either result. The reconciliation definition explains why agreeing calculations are not enough when their underlying populations differ.

Record acceptance and limitations

Keep the control sheet, source filename, export cutoff, definition version and review date together. Check identifier uniqueness at the source's stated level, required fields, sign conventions and missing dates. When totals are expected to match exactly, any difference needs explanation; a materiality threshold should never silently authorize missing records.

The business owner should approve the intended use after reviewing the evidence. If a month is still open, label it provisional and avoid comparing it with a complete month as though both are final. Track unresolved exceptions with an owner and a next step.

Use dashboard software selection guidance when evaluating tools. This acceptance standard answers a separate question: whether a particular set of figures is trustworthy enough for its stated business use.

Where dashboard standards go wrong

  • A familiar label hides a changed definition. Revenue, active customer or on-time delivery can change meaning while the chart title stays the same. Record versions and effective dates.
  • Periods look equal but are incomplete. A current month through yesterday cannot be read as a full-month comparison. State the cutoff and comparison basis.
  • A total agrees while segments overlap. Confirm that category or department assignments are mutually exclusive where the report expects them to add up.
  • Review evidence disappears. An approval without the source, calculation and exception record cannot explain a later discrepancy.

Covirage is not affiliated with Microsoft.

Check your own dashboard figures in Covirage

Upload the source export and documented measure definitions to Covirage. Covirage maps the columns, its tools compute and check the figures, and the external AI model explains the results with every figure traced to its rows. Create a free Covirage account, which takes about a minute, then set up your business and bring a sample of your own data.

See metrics governance in Covirage for the supporting business context.

Questions people ask

What belongs in a dashboard acceptance standard?

Record each measure's name, formula, unit, source, inclusion rules, date basis, comparison period and owner. Add reconciliation checks and the evidence required for approval. The standard should let another person reproduce the result without asking the dashboard author to interpret it.

How do you validate a dashboard total?

Recalculate it from the approved source rows under the documented filters. Compare that amount with the dashboard and the source control total. Investigate any difference, then record whether the cause was a filter, missing record, duplicate, timing difference or definition mismatch.

Who should approve dashboard measures?

The business owner approves the definition and intended use. The source owner confirms completeness and timing, while the preparer demonstrates the calculation and reconciliation. The approving person should inspect the evidence rather than treating a plausible chart as proof.