Prompt status and editing
Use the prompts list to review questions and classifications and manage what you continue to measure. Status tabs filter the list. Selecting a tab does not change a prompt's status.
Manual-only enabled prompts
The Enabled tab appears between Draft and Scheduled. Enabled prompts can run on demand but are excluded from scheduled measurement. Select Enabled from a draft's status menu to make it ready for manual runs. Running a draft on demand moves it to this manual-only state, without enabling a schedule.
The existing API and CLI value status: disabled represents this manual-only state; active enables scheduled measurement. Saved configuration and previous answers are preserved.
What the status tabs mean
| Tab | What it includes | Scheduled measurement |
|---|---|---|
| Suggested | Questions not yet accepted for tracking | Excluded until accepted and enabled |
| Draft | Prompts being prepared | Excluded |
| Enabled | Prompts enabled for manual runs only (disabled in the API) | Excluded |
| Scheduled | Prompts enabled for scheduled measurement (active in the API) | Included |
Archived prompts are shown separately through the archive view at the bottom of the sidebar. Tab counts count prompts, not answers, executions, or mentions. Search, topic, and other filters can make the visible row count differ from the tab count.
Active
The prompt is eligible for scheduled measurement. Active describes its configuration, not whether the latest execution succeeded. A newly added prompt may still be waiting for its first answer. Check answer counts and the collected answers to review results.
Enabled (manual only)
Use this status for on-demand runs without a schedule. Previous answers and configuration are preserved. Choose Enable scheduled measurement separately to start automatic runs.
Suggested
These questions are not yet tracked. Review their intent, audience, and topic before accepting them. They become measurement targets after successful acceptance and activation. You can reject unwanted suggestions. See Set up prompts for question design and creation.
Archived
Archived prompts are set aside from future observation. Archiving and deleting are separate actions.
Currently, the normal Enable action cannot resume an archived prompt. Use Disable if you plan to resume it later.
Change a prompt's status
- Find the prompts using status tabs or search.
- Select their row checkboxes.
- Choose Disable, Enable, or Archive from the selection actions.
- If a confirmation appears, review the targets and action. Check the completion message and destination tab for the result.
A bulk action may succeed for only some prompts. Review failed items before retrying. Switching tabs or a row disappearing from the list is not enough to establish that an action succeeded.
Edit classifications and measurement settings
Use Branding, Consideration stage, and Tags in the list to review and edit classifications. Select multiple prompts to change classifications, tags, topic, region, language, or platforms together.
| Field | Purpose |
|---|---|
| Branding | Classify whether the question names a brand |
| Consideration stage | Classify the stage: Information gathering → Research & comparison → Purchase & sign-up |
| Topic and tags | Organize questions and narrow the analysis scope |
| Region, language, and platforms | Specify the conditions used to measure AI answers |
Classifications organize questions. Changing a branding label or tag does not itself increase brand mentions in answers. A change to region, language, or platforms changes the measurement conditions; compare results under consistent conditions.
If you want to track a different intent by changing the question, add a new prompt to keep it distinct from the earlier question. Use Open on a prompt to review its details and answers.
Distinguish status from performance
Status describes whether a prompt will be measured; metrics describe what happened in collected answers. Disabling a prompt does not turn its past visibility into 0%. An active prompt without answers does not yet have evidence of performance.
When comparing results, check the period, prompts, AI models, and answer counts. Adding or stopping prompts, or changing filters, can change the sample composition and affect aggregate metrics.
- Main metrics: review visibility, average position, mentions, and answers
- Organize with topics and tags: use classifications and filters
- Answers: read the collected answer text