Validation report

The output of the checks run on a file before any figure from it is shown: what parsed, what is missing, what is duplicated, and whether the roll-up reconciles, with the rows behind each line.

DefinitionThe output of the checks run on a file before any figure from it is shown: what parsed, what is missing, what is duplicated, and whether the roll-up reconciles, with the rows behind each line.

What it contains

One line per check: pass or fail, the figure, and the detail. "Every value in Net Sales is a number, 290 rows." "1 row has no region, blank Region, row 290." "Each account sits under one person, 289 accounts." "region = person = account = total, 1,403,577, 0.00 variance."

Why it comes first

The report is the thing the visitor sees before the dashboard, because a dashboard drawn on a file that does not reconcile is a dashboard that will be contradicted in the meeting. A report that says what is wrong, by row, is the product's first useful output.

Same report, any door

A file uploaded in the browser, posted to an API, or dropped in storage on a schedule gets the same report back.

How it is used

Before any figure is shown, the file is checked: rows parsed, columns mapped, dates valid, identifiers matched, totals against the control total, duplicates. The report lists what passed, what failed and what was excluded.

Example

41,280 rows read; 41,102 parsed; 178 excluded for an unreadable date, worth $61,400; total matches the control total once those are added back.

Where it goes wrong

Bad rows dropped silently. The totals are slightly low and nobody knows why.

Articles that use this term

AI and self-service analytics

Asking questions of a sales workbook without rebuilding it in a BI tool

How a team whose data lives in Excel or Google Sheets can ask plain-English questions of the workbook they already keep: reading the structure as it is, mapping columns once, validating the totals, and getting answers with the rows behind them, without pivot tables or a data warehouse.

16 Sept 20263 min read Data quality and reconciliation

Control total and identity: the check against the world, and the check against itself

The difference between a control total, which tests an uploaded file against the figure the source system showed, and an identity, which tests the computed tables against each other, why both are needed and neither replaces the other, the failure each catches that the other cannot, a filtered export against a duplicated account, the order they run in, and what a report says when one of them is unavailable.

16 Sept 20262 min read AI and self-service analytics

Deterministic first: the architecture of an analytics product that uses a model and stays right

How an analytics product uses a language model without letting it touch a number: deterministic code computes every measure from stated definitions, a tool registry names each computation, the model plans which tools to call and reads their outputs, every output carries its citation, and the model's own text is labelled as a reading. The four boundaries that make this hold, what crosses each, and what never does.

16 Sept 20263 min read Wallet share and penetration · Asset managers

How to calculate share of wallet for an asset manager's distribution team

A working method for share of wallet in asset management distribution: what counts as the wallet, which rows you need from your own systems, the roll-up level by level, and the three places the number usually goes wrong.

16 Sept 20264 min read Coverage and territory · Investment banking

Sales coverage on an institutional desk: measuring it by salesperson, client and product

How an investment bank's institutional sales desk measures coverage from its own revenue and contact data: which clients each salesperson reaches, which products each client trades with the desk, and the roll-up that reconciles to the benchmark submission.

16 Sept 20264 min read Data quality and reconciliation

Ten questions a head of sales operations asks, and the check that answers each

The ten questions a head of sales operations asks before anything is reported, does the CRM reconcile to the ledger, does every account have one owner, do the territories' quotas sum to the target, are the splits summing to one, is the pipeline clean, has any identifier drifted, which reps' data can be believed, did the definitions change, did the export match its control total, and what is on the exception list, each with the check that answers it, the identity behind it, and why sales operations owns the checks that everyone else's numbers rest on.

16 Sept 20263 min read