Define Meaningful Change Rules

Why this decision matters

A fingerprint is a cheap signal that something may have changed. A field comparison explains what changed. A significance rule determines whether the change belongs in the report. These are separate stages. Noise filters can exclude navigation, copyright dates, rotating testimonials, or tracking parameters, but each filter creates a risk of hiding real evidence and therefore needs a fixture. Removals deserve special treatment because a missing element may reflect redesign or fetch failure.

This lesson advances the same capstone used throughout the course: the Competitor Intelligence Monitor Pack. The learner is not collecting ideas for later. The workbook record created here becomes an input to the next module and must be specific enough that another operator could review it. Where evidence is incomplete, the artifact should show the gap instead of smoothing it over.

Source evidence and limits

C6-S3 shows fingerprints, product-field comparisons, first-run handling, duplicate controls, block-page detection, shell detection, and typed add, remove, product-change, and status-change events.

The sources support the operating method stated above. They do not prove universal performance, guaranteed savings, complete market coverage, causal impact, or a client result. Any date-sensitive product, platform, competitor, legal, pricing, or policy fact must be checked against a current primary source before publication or operational use.

Continuing fictional example

Example: HarbourOps is a fictional Toronto workflow consultancy monitoring three invented competitors: BeaconFlow, TaskHarbor, and RelayDesk.

BeaconFlow's footer year changes and is ignored. Its Starter tier changes from a fictional monthly amount to contact sales; that field change is retained for review. TaskHarbor returns a loading shell, so no pricing removal is emitted.

Example boundary: Every company, page, event, and result in this example is fictional. It demonstrates the method and is not observed market evidence.

Workbook application

Use Workbook Section 4: Define Meaningful Change Rules. Complete Workbook Section 4. Define at least six event types, the fields compared, ignored regions, confidence rule, reviewer action, and one fixture per rule.

1. Write the current evidence or input before adding interpretation.
2. Apply the lesson's decision rule and state the reason for the classification.
3. Mark uncertainty, missing information, and the human owner for the next decision.
4. Check that the result stays inside the course boundaries and can be tested.

Failure modes to inspect

Alerting on every hash; interpreting an empty response as a discontinued product; deduplicating different events because their titles match; and using an LLM interpretation without retaining the deterministic observation.

A polished output can still fail if its evidence, permissions, identity, denominator, source date, or action boundary is wrong. Review the underlying record rather than grading tone alone. The correct repair may be to narrow the scope, gather a permitted source, label an unknown, or stop the proposed action.

Decision rules

Fetch success precedes comparison. Normalization precedes fingerprinting. Field evidence precedes interpretation. Materiality belongs to the named business decision, and uncertain removals require verification.

Record the rule in the workbook in language two reviewers can apply consistently. A rule that depends on intuition alone cannot support a deterministic fixture. If reviewers disagree, preserve both readings, identify the missing evidence, and revise the rule before automation.

Three application checks