Recording a visibility baseline
Check the brand and measurement scope, then save visibility, citations, and sentiment for the next comparison.
When to use this workflow
“Show our current strengths and weaknesses across AI platforms.”
Before fetching
Fix brands, competitors, platforms, dates, and the prompt cohort. Return a comparable baseline and the first prompts to investigate.
Quick reference
Use this sequence once you know the task. Read the numbered steps below for the reasoning and checks.
orc brands list --json
orc prompts list --json
orc reports visibility get --json > visibility.json
orc reports citations get --json > citations.json
orc reports sentiment get --json > sentiment.jsonPrerequisites
Complete the CLI quickstart. These commands use CLI 0.5.0. Select the target workspace and use IDs returned by list commands.
Collected answers and reports for the target brand.
Check the brand and tracked questions, then save the three reports. Record the workspace, brands, questions, AI platforms, regions, returned reporting period, and retrieval time with the files.
1. Confirm the brand and competitors
Check the target brand ID and tracked competitors. Use the ID when filtering answers.
orc brands list --json2. Inspect tracked questions
Read question text and returned settings, then select the prompt IDs for this investigation. Fetch subsequent pages when needed.
orc prompts list --json3. Fetch visibility
Check the returned period and dimensions. Choose a baseline with matching conditions and keep mismatched data separate.
orc reports visibility get --json > visibility.json4. Fetch citation observations
Inspect sources cited in answers. Keep citations distinct from crawls and actual visits.
orc reports citations get --json > citations.json5. Use sentiment to narrow the investigation
Use changes to select answers for inspection. A score alone does not determine whether a claim is wrong.
orc reports sentiment get --json > sentiment.json6. Verify and save the result
Check each command's exit status before using its file. The result is a baseline you can revisit under the same conditions. If a report does not return a brand or platform breakdown, it cannot establish differences within that breakdown.
Choose channels and prompts to investigate
Fetch platform visibility under matching conditions and compare stronger and weaker channels. Fetch comparison brands separately with --dimensions brand; share of voice and visibility are distinct metrics.
orc reports visibility get --dimensions platform --metrics visibility_rate,mention_count --date-range '{"period":"7d"}' --json
orc reports visibility get --dimensions brand --metrics share_of_voice --date-range '{"period":"7d"}' --jsonHave the agent inspect answer counts and scope, then select prompts behind the largest differences. Follow with an answer audit rather than inferring causes from aggregates.
When data is missing
If data is missing, check the workspace, collection period, filters, and pagination. Keep a failed request separate from an empty result. Use --help to inspect supported options.
If a failure remains, report it and verify the fix, including this workflow and the failed step.