Skip to main content

​
Direct answer

Use Cockpit > Compare with one funnel stage at a time. Identify where DIRECT competitors have more observed presence, then trace the difference to prompts and answers. Define what losing means before ranking stages.

​
Data required

  • Answered prompts with consistent funnel-stage assignments
  • Correct SELF and DIRECT competitor relationships
  • Matched provider, model, country, language, topic, tag, and period filters
  • Mention Rate, Coverage, Share of Voice, position, and their raw components
  • Full answers for the prompts contributing to each gap

​
Workflow

  1. In Workspace > Prompts, review missing or inconsistent funnel-stage assignments.
  2. In Cockpit > Compare, select a funnel stage and one metric.
  3. Record the SELF and DIRECT competitor values with their raw counts.
  4. Repeat for the other stages without changing the remaining filters.
  5. In Cockpit > GEO Matrix, locate the topic and prompt cells behind the largest observed difference.
  6. Read those answers in Monitor > Response Analysis to check recommendation context and relevance.

​
Interpretation

Product Share of Voice is occurrence-level, Product Mention Rate is response-level, and Product Coverage is prompt-level. They can identify different stages as the largest gap. Do not combine them without an approved method. A stage with a small answer sample may show a large percentage difference. A competitor mention can be negative or incidental, so answer context determines whether the observed gap is strategically relevant.

​
Possible next actions

  • Test a content change against the specific prompt and decision criterion behind the gap.
  • Improve funnel-stage assignments when taxonomy quality explains the difference.
  • Investigate one provider or topic instead of applying a site-wide response.
  • Track the same prompt cohort after the experiment.
These actions test hypotheses and do not guarantee future provider behavior.

​
Limitations

  • Funnel-stage assignment depends on your taxonomy.
  • Competitor metrics depend on DIRECT relationship configuration.
  • Unequal samples can distort stage comparisons.
  • Visibility differences do not establish commercial loss or causality.