Skip to main content

​
Direct answer

Use Cockpit > Overview to compare a stable prompt, provider, model, topic, and competitor cohort before and after the campaign. Add answer-level evidence and record any concurrent changes. Describe movement as associated with the campaign period unless you have a separate causal design.

​
Data required

Use campaign start and end dates, unchanged configured prompts, providers and models, topics or tags, SELF/DIRECT classifications, stored answers and citations, comparable periods, and a log of content, PR, product, competitor, and monitoring changes.

​
Workflow

1

Define the baseline

Choose a pre-campaign period and record the exact prompt, provider, model, topic, geography, and competitor configuration.
2

Mark the campaign window

Record launch, publication, distribution, and end dates. Note which pages, claims, and audience topics the campaign targeted.
3

Compare the stable cohort

Review Mention Rate, Citation Rate, Share of Voice, Coverage, sentiment, and representative answers only where they answer the campaign question.
4

Audit alternative explanations

Check model, prompt, source, competitor, product, and tracking changes before interpreting the difference.

​
Interpretation

Product Mention Rate, Citation Rate, Share of Voice, and Coverage use distinct units and denominators; retain their documented definitions. A post-campaign increase is temporal association. It does not establish that the campaign caused provider outputs to change.

​
Possible next actions

  • If movement is concentrated in targeted topics, test the hypothesis with additional comparable runs before extending the campaign.
  • If citations change without SELF mentions, inspect the answers and cited URLs rather than calling the campaign successful.
  • If no change is observed, verify cohort stability and source availability before revising the campaign.

​
Limitations

Provider outputs, models, web access, sources, competitors, and monitoring configuration can change during the comparison. Qwairy observes selected stored answers and cannot isolate campaign causality or guarantee a response timeline.

Exports

Export the comparison cohort.

Performance Overview

Build a period comparison from documented data.

Monitoring

Review the configured prompt set.

Visibility over time

Apply stable-cohort comparison rules.

Content pickup observations

Track crawler and citation events separately.

Share of Voice

Use the product occurrence formula.

Automated reporting

Create a repeatable reporting workflow.