No data for an engine

AmICited tracks eight engines: ChatGPT, Perplexity, Gemini, Google AI Overview and Google AI Mode on every plan, plus Claude, Grok and DeepSeek on Premium and Enterprise. When one of them shows no data, there are two very different situations: the engine did not run, or it ran and did not mention you. The dashboard can make them look the same, so the first step is always to find out which.

First, did it run?#

Open any prompt from Prompts and look at its detail. Each engine shows its answer, or “not run” / “not run yet” if it has no answer in the period. The header also says how many engines cited you, for example “cited by 2/5 engines”.

• Answers exist but do not name you: the engine ran. Skip to cause 5.
• “Not run” on every prompt: work through causes 1 to 4.

1. It is a premium engine and your plan does not include it#

Claude, Grok and DeepSeek are only queried for Premium and Enterprise workspaces. On other plans they are skipped on every run, including after a downgrade from Premium, even if prompts were set to use them.

• Check. Your plan on pricing or in the workspace billing settings.
• Fix. Upgrade to Premium to add them. See AI engines and plans and limits.

2. The prompt’s engine list excludes it#

Each prompt can have an engine allow-list. Empty means every enabled engine (and new engines are added automatically). If someone picked specific engines, the others never run for that prompt. The same applies to the AI providers you selected during onboarding.

• Check. Edit the prompt on Prompts and look at its engines.
• Fix. Add the engine back, or clear the list so the prompt uses all engines.

3. It has not run yet#

New prompts fill in engine by engine, and a weekly or monthly schedule means an engine may have one answer or none in a short window.

• Check. The prompt’s status (new, pending and processing mean runs are still in progress) and its schedule.
• Fix. Wait for the next run, or widen the date range.

4. Runs are failing or paused#

A failed run is retried after 5 minutes. A prompt whose last run failed shows a failed status. If the workspace is out of credits, the domain is paused and no engine runs at all.

• Check. In the prompts table, filter by Failed. On the Dashboard, look for “Your domain is inactive”.
• Fix. Top up credits. If an engine keeps failing on the same prompt, contact us with the prompt.

5. It ran but never mentions you#

This is the most useful finding on this page, and the easiest to miss: the By platform breakdown on the dashboard only lists engines that mentioned you. An engine that ran on every prompt and never named you does not appear at all, so the breakdown suggests you are visible “everywhere” it shows.

• Check. Filter the Dashboard to that engine: visibility 0% with prompts that have answers confirms it. In prompt detail, read that engine’s answers and sources.
• Fix. Engines differ in where they look. Google AI Overview and AI Mode lean on Google’s index; ChatGPT and Perplexity search the web in their own way. Use Sources filtered to the engine to see which sites it trusts, and target those. The ai-visibility-audit skill guides an agent through this per engine.

Engines that are not tracked#

Microsoft Copilot is not tracked. On the Bing Webmaster Tools side, Bing query data is the closest read on Copilot demand; see Search Console and Bing. Raw Google search results are stored alongside AI Overview as “Google Search” but are not an AI engine.

Related#