Workflows / AI search visibility
30 buyer prompts asked to 3 assistants, 90 answers with their citations and share of voice for 4 brands.
Give this to your agent
Using RouterGrowth, run this workflow: Check where ChatGPT, Claude and Perplexity mention our agency across 30 prompts and find the gaps versus 3 competitors. Show me the price before every call, and stop after each stage with a sample of the results so I can adjust. 1. Pick the prompts: aeo.keywords x 3 searches. What people actually ask assistants in the category, with volume. 2. Ask the assistants: aeo.answer x 90 (30 prompts x 3). Each prompt against 3 assistants, answers with citations. 3. Track the mentions: aeo.mentions x 4 brands. Share of voice for you and the competitors you named. 4. Read what gets cited: web.scrape x 60 cited pages, seo.serp x 30 (Google comparison). The sources assistants lean on are the pages to beat or to be on. Budget for one run: about $2.78. Run routergrowth inspect on any capability before the first call to confirm the quote.
Works in Claude, ChatGPT, Claude Code, Codex and Cursor. New workspaces start with $1 of credit. Get a key or read the setup guide.
Why it works
A buyer asking an assistant for an agency gets names drawn from directories and case studies. The gap is usually a listing, not a ranking.
Buyers now ask an assistant before they search. Whether it names you, and what it cites when it does, is a ranking you can measure. These workflows pick the prompts people actually use, ask every assistant, track mentions for you and your rivals and read the sources that get cited so you know which pages to be on.
What the run hands back
The agencies named per prompt, the directories cited and the ones that do not list you.
The workflow
4 stages and 187 calls on one key. Every call is quoted and reserved before it runs, and a lookup that finds nothing releases the hold.
The sources assistants lean on are the pages to beat or to be on.
The receipt
About $2.78 for the whole run. No subscription and no seats: each call is paid from a prepaid balance at the price shown before it runs.
| Capability | Calls | Cost |
|---|---|---|
| aeo.keywords$0.0177 per search | 3 | $0.05 |
| aeo.answer$0.0175 per answer | 90 | $1.58 |
| aeo.mentions$0.1767 per search | 4 | $0.71 |
| web.scrape$0.0056 per page | 60 | $0.34 |
| seo.serp$0.0035 per search | 30 | $0.11 |
| One run | 187 | $2.78 |
Assumptions: 2 contacts per account unless the brief says otherwise, a 70% email find rate, 90% of found addresses passing verification, 3 emails per contact and a 30% LinkedIn accept rate. Prices are the catalog's list prices when this page was built. Change the volumes and the assumptions in the workflow generator, or see how plans lower the price of every call.
# RouterGrowth workflow: Check where ChatGPT, Claude and Perplexity mention our agency across 30 prompts and find t
npm install -g routergrowth && routergrowth setup
# 01 Pick the prompts (aeo.keywords x 3)
routergrowth run -c aeo.keywords -i '{}'
# 02 Ask the assistants (aeo.answer x 90)
routergrowth run -c aeo.answer -i '{"prompt":"<prompt>"}'
# 03 Track the mentions (aeo.mentions x 4)
routergrowth run -c aeo.mentions -i '{}'
# 04 Read what gets cited (web.scrape x 60, seo.serp x 30)
routergrowth run -c web.scrape -i '{"url":"<url>"}'
routergrowth run -c seo.serp -i '{"keyword":"<keyword>"}'
FAQ
aeo.answer asks the assistant a prompt with web search on and returns the answer with every source it cites. Run it across the prompts your buyers use and count where you are named.
ChatGPT, Claude, Gemini and Perplexity for answers. aeo.mentions adds where a brand or domain appears in Google AI Mode and ChatGPT across a wider set of questions.
aeo.keywords shows how often questions are asked to AI assistants, with a monthly trend. Start from the questions with volume in your category and add the ones your sales team hears.
Read the cited sources. The pages an assistant leans on for a prompt are the pages to be mentioned on or to outdo, and that list is the plan.
About $2.78 for one run of 187 calls. Each call is quoted before it runs and a lookup that finds nothing is not charged, so the figure is a ceiling. There is no subscription: the run is paid from a prepaid balance.
Keep going
Reading this as an agent? This workflow as markdown: /workflows/aeo-agency.md · every workflow: /workflows.md
Paste the prompt into your agent. The first calls run on the $1 that comes with signup.