BizHallu: business risk and metric scope

One ledger, explicitly different scopes

December 2010 through December 9, 2011; December 2011 is partial. 541,909 raw records; 534,129 eligible ledger lines under the historical cleaning policy.

GBP 10,642,110.80 + (GBP -893,979.73) = GBP 9,748,131.07

Net merchandise value is GBP 9,771,708.99. It exceeds the all-ledger net here because the other categories have a negative net contribution; these are different scopes, not competing estimates of the same population.

46.77% of the absolute negative value is in non-merchandise categories under that heuristic: GBP 418,078.57 / GBP 893,979.73. This is not a physical return percentage or proof that those records are bad data.

Preserve existing valid-net flags: exclude normalized duplicate copies, missing/blank descriptions and nonpositive unit prices. Missing customer IDs are retained for aggregate analysis, not imputed. No anomaly is automatically removed.

UCI documents C-prefixed invoices as cancellations. Negative quantity alone does not establish a physical return, cause, original-sale linkage, cash collection, profit, or recognized accounting revenue.

Transaction-value reconciliation

Accounting and reporting

Do positive and negative values reconcile to the reported net for the same period and scope?

Control: Check the sign, period, unit, eligibility policy and additive identity before interpreting change.

Boundary: Net transaction value is not audited financial-statement revenue, profit or cash collected.

Negative-value review

Accounting and operations

How much negative value comes from merchandise versus fees, adjustments and unknown codes?

Control: Keep sign, invoice cancellation flag and merchandise classification separate. Do not allocate unlinked adjustments to products.

Boundary: Negative merchandise lines are not an original-sale-linked physical return rate; cause and timing remain unknown.

Product concentration and ranking

Supply management and performance reporting

Are product, period, metric, rank and value supported by the same evidence scope?

Control: Check complete relationships and the concentration denominator. Preserve stock-code-plus-description grain for the historical study.

Boundary: Sales-value concentration does not establish inventory requirements, demand forecasts, procurement savings or optimal replenishment.

Country exposure and comparison

Business and operations analysis

Are the named countries, comparison direction and amount difference consistent?

Control: Subtract within the same month and scope; distinguish domestic/non-domestic exposure and disclose missing dimensions.

Boundary: Transaction geography alone cannot explain market attractiveness, profitability, customer acquisition or causal growth drivers.

From a business claim to an evidence check

In the April example, the product's own amount is correct while its stated rank is wrong. The September example shows three such ranking mismatches. These are curated examples, not estimates of business-error frequency or loss.

A low-uncertainty list marker does not establish confidence in the later product-rank-amount relationship. The methods comparison reports exploratory uncertainty signals alongside simple references and limitations.

What this project has not measured

No avoided loss, analyst time saving, inventory benefit or deployed control effectiveness has been measured. Suggested use is a review control, not an implemented business decision system. Row counts and question coverage do not establish impact.

A useful next assessment would compare these evidence checks with deterministic SQL and template-based reporting, then measure where review adds value. The present examples establish an auditable question, not a quantified operational benefit.