Forecast evaluation / Delivery leads, planning owners and acceptance reviewers

Troubleshooting path

Keep zero outcomes out of undefined percentage-error claims

Inspect a forecast error when its observed-value denominator is zero.

This checks an observed-zero denominator in percentage error, distinct from a zero training-reference scale.

Request TeamBoostAI access Get the work brief Task table (.csv)

Opens the current invite-request page. Access is subject to approval; this example is not imported automatically.

Original worked case · Manual planning resource

Inspect the decision, not just the summary.

Start with the invented evidence, follow the reasoning, and retain its limits when adapting the brief.

02 · Follow the reasoning

How the case leads to a decision

Absolute percentage error uses 100×|actual−forecast|/|actual|. B gives 100×1/4=25%; A requires 2/0 and is undefined. A’s unscaled absolute error remains two minutes.

The bounded result

A’s absolute percentage error is undefined because its observed denominator is zero. B’s is 25%; report A’s two-minute absolute error separately rather than inventing an overall finite percentage average.

01 · Inspect the inputs

Every record stays visible

Inspect the invented case records
Case	Forecast recorded minutes	Observed recorded minutes	Absolute error	Absolute percentage error
A	2	0	2 min	Undefined: zero denominator
B	3	4	1 min	25%

Use horizontal scrolling for wide tables. These records are invented, not customer data.

Records · CSV Complete case · JSON Manual brief · Markdown

Definitions and method context

Records, decisions and policies are original synthetic examples. External references supply context; they do not validate these cases or TeamBoostAI capabilities.

Forecasting: Principles and Practice — accuracy ↗
Genuine held-out forecast and point-error evaluation context. Binary scoring examples use their own explicit toy definitions; no real task model is validated.

External references checked 6 October 2026. Demand for these topics has not been measured.

Inspect the conditions behind this decision

Diagnostic planning path

Synthetic planning case: A toy duration evaluation has A:forecast two recorded minutes, observed zero; B:forecast three, observed four. A’s zero can reflect the recording convention, whose physical meaning is unresolved.

Verify observed-value semantics

Owner role · Evaluation owner

Evidence: Zero’s recording convention and unit are retained without inferring instantaneous physical work.

Keep undefined scores visible

Owner role · Reviewer

Evidence: A is not replaced with zero percent or silently dropped from all-case coverage.

Choose a justified comparison

Owner role · Planning lead

Evidence: Report compatible unscaled errors or an explicitly justified different metric, with its changed meaning declared.

Decision to make: A’s absolute percentage error is undefined because its observed denominator is zero. B’s is 25%; report A’s two-minute absolute error separately rather than inventing an overall finite percentage average.

Filled manual planning note

Invented planning text. Adapt it to your evidence and confirmed owners.

A’s absolute percentage error is undefined because its observed denominator is zero. B’s is 25%; report A’s two-minute absolute error separately rather than inventing an overall finite percentage average. Absolute percentage error uses 100×|actual−forecast|/|actual|. B gives 100×1/4=25%; A requires 2/0 and is undefined. A’s unscaled absolute error remains two minutes. Recorded zero does not establish actual effort or duration semantics. No future forecast performance is inferred.

Questions about this workflow

Does a zero observed value mean zero prediction error?

No. A predicts two while zero is recorded, so its absolute difference is two.

Can I divide by a tiny substitute value?

Not silently. That introduces an arbitrary scoring rule and can radically change the result.

Does this create tasks in the product?

No. The brief is a manual planning resource. Use the product access link to check onboarding and the workflows available in your account.

Put the outline to work

Verify observed-value semantics. Check: Zero’s recording convention and unit are retained without inferring instantaneous physical work.
Keep undefined scores visible. Check: A is not replaced with zero percent or silently dropped from all-case coverage.
Choose a justified comparison. Check: Report compatible unscaled errors or an explicitly justified different metric, with its changed meaning declared.

From a useful outline to team work

Explore TeamBoostAI for your team

Use the owned checks and downloaded brief to discuss this planning decision alongside your TeamBoostAI tasks. Confirm available fields, roles and account features separately. The example is manual; it does not calculate live analytics, create work or run an experiment in the product. Confirm the workflows available in your account before adopting this outline.

Request TeamBoostAI access