Edcen Data All articles
Data Governance

What Went Into That Figure? A Provenance Note for Every UK Dashboard

Edcen Data
What Went Into That Figure? A Provenance Note for Every UK Dashboard

UK boards are now used to seeing a number on a screen and treating it as settled fact. A dashboard says customer churn is 4.2 per cent; an AI tool flags a supplier as high risk; a finance pack says margin improved. The figure arrives with a colour, a trend line and an air of authority. What it rarely arrives with is a short explanation of how it was made.

That gap matters more in 2026 than it did when reporting cycles were slower. Mid-market organisations often run analytics with small teams, multiple source systems and a growing list of AI-assisted outputs. The answer is not necessarily a heavyweight data catalogue. A simpler artefact can do much of the work: a one-page provenance note attached to every key metric.

The trust gap on the dashboard

Consider a familiar scene. A monthly metric moves sharply. An executive asks why. An analyst disappears into spreadsheets, queries and job logs to reconstruct the path from source system to dashboard. Two days later, the answer arrives with an apology: a filter was changed, a manual adjustment was applied, or a source extract was late. The number was not wrong, but nobody could explain it quickly.

A provenance note prevents that scramble. It is not a technical lineage diagram, although it should align with one. Technical lineage maps tables, jobs and transformations for engineers. A provenance note translates the same journey into plain English for decision-makers, auditors, new starters and anyone who has to defend a figure outside the analytics team.

For UK businesses, the value is practical. Boards want assurance. Regulators and customers expect explanations. Finance teams need to reconcile. Regional firms with lean data capability need auditability without building a catalogue programme that consumes the year.

What the note should contain

The note should fit on one page and use language a non-specialist can follow. A useful template covers seven fields.

Field Plain-English prompt
Metric name and definition What exactly is being counted, and what is excluded?
Source systems Which systems or files feed the number?
Transformation steps What happens between source and figure, including manual adjustments?
Owner Who is accountable for the metric and its note?
Last validation date When was the figure last checked against source or a known control?
Known caveats What should users know before acting on the number?
Review cadence When will the note and metric next be checked?

A note for a metric such as monthly active customers might say that it draws from the CRM and billing platform, excludes internal test accounts, applies a thirty-day activity rule, and was last validated on 31 October 2026. It might name the commercial operations manager as owner and note that a recent CRM migration has made October data less reliable than usual. That is enough to stop a wrong decision or an unnecessary escalation.

Writing rules that keep it useful

The first rule is brevity. A note that runs to five pages will not be read. The second is plain English. Avoid phrases such as 'various sources' or 'business logic applied'. Name the system, the rule and the person. Use absolute dates rather than 'last quarter'. State caveats directly rather than burying them in a caveat section nobody opens.

Ownership matters. A team name is not an owner. A named role, backed by a named individual, creates accountability for validation and review. The note should also carry a version number and a last-updated date, so users know whether they are looking at current documentation.

Not every metric needs a note on day one. Start with the numbers that appear in board packs, regulatory returns, customer-facing reporting and AI inputs. Those are the figures where a provenance gap carries the greatest cost.

A ninety-day rollout for lean teams

A mid-market firm can introduce provenance notes without buying a new platform. Weeks one and two: select ten high-value metrics and agree the template. Weeks three and four: assign owners and draft the first notes with analysts and finance. Weeks five and six: review the notes with the people who use the metrics, not only the people who build them. Weeks seven and eight: publish the notes alongside the metrics in the existing BI tool, wiki or reporting pack. Weeks nine to twelve: extend to AI inputs and outputs, then set a quarterly review cycle.

The rollout should be deliberately modest. The aim is a habit, not a documentation project. If the note cannot be found from the dashboard in two clicks, it will not be used. If the owner cannot be identified, the metric is not ready for Board-level reliance.

Governance without a catalogue project

A provenance note is not a substitute for data governance. It is an entry point. It creates accountability, surfaces known weaknesses and gives auditors and new joiners a route into the metric. Over time, the notes can feed a fuller data catalogue or lineage tool. They can also support AI governance by showing which inputs and assumptions sit behind a model output.

For UK organisations with limited resources, the attraction is proportionality. The note does not require a platform migration or a new team. It requires a template, a named owner and a review date. That is governance at the point of use, which is often where trust is won or lost.

A number without provenance is an assertion. With a short note, it becomes evidence. UK firms do not need a perfect catalogue to start. They need a simple habit: before a figure reaches the board, someone should be able to answer, in plain English, what went into it.

All Articles

Related Articles

What the Timestamp Doesn't Tell You: A Freshness Audit for UK Data Teams

What the Timestamp Doesn't Tell You: A Freshness Audit for UK Data Teams

The Answer Engine: Why UK Data Catalogues Need to Start With Questions, Not Tables

The Answer Engine: Why UK Data Catalogues Need to Start With Questions, Not Tables

The Kilowatt Ledger: Bringing Energy into UK Analytics Governance

The Kilowatt Ledger: Bringing Energy into UK Analytics Governance