400 public posts read and 120 authors profiled; then 120 accounts and 120 people, 120 LinkedIn connection requests and about 76 verified work emails for the follow-up.
Give this to your agent
Using RouterGrowth, run this workflow: Mine Reddit and LinkedIn for people complaining about their applicant tracking system, 400 posts, then email the head of people at their companies. Show me the price before every call, and stop after each stage with a sample of the results so I can adjust. 1. Find the conversations: social.search x 400 posts. Public posts matching the pain, across X, Reddit, LinkedIn and the rest. 2. Read the threads: social.comments x 1,200 comments, social.profile x 120 authors profiled. Comments carry the objections and the names; profiles tell you who is worth a look. 3. Attach the companies: company.enrich x 120 companies. Authors to companies: firmographics on the domain behind each profile. 4. Get the contact: contact.find x 84 (120 tried, 70% found), contact.verify x 84 verdicts. Charged on a match only: a miss is free, so the estimate is a ceiling. 5. Context for the first line: web.scrape x 120 pages, social.posts x 360 posts. One page per account and a few recent posts per person, so the opener is about them. 6. Run the outreach: email.send x 228 (76 contacts x 3 steps), linkedin.invite x 120 requests, linkedin.message x 36 (30% accepted). Connection request first, email to the rest; both from accounts you own. 7. Read the replies: email.messages x 23 inbox reads, linkedin.messages x 12 inbox reads. Poll the inboxes, classify replies, hand the warm ones to a human. Before sending: email.domain x 1, email.inbox x 1, on a domain I own. Budget for one run: about $9.65 plus $4.66 one-time setup. 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
Recruiters complain about their ATS to each other, in detail, with the vendor named. That thread is a requirements document written by the buyer.
People say in public what they would never put in a form: the tool they are leaving, the feature that broke, the vendor that ignored them. These workflows find the conversations, read the threads, profile the authors worth a look, attach their companies and reach out with the complaint as the opener.
The first line, as your agent would write it
Your comment about candidates disappearing between stages is the most common ATS complaint I read this month. Which stage loses them?
The workflow
7 stages and 2,907 calls on one key. Every call is quoted and reserved before it runs, and a lookup that finds nothing releases the hold.
Public posts matching the pain, across X, Reddit, LinkedIn and the rest.
Comments carry the objections and the names; profiles tell you who is worth a look.
Authors to companies: firmographics on the domain behind each profile.
Charged on a match only: a miss is free, so the estimate is a ceiling.
One page per account and a few recent posts per person, so the opener is about them.
Connection request first, email to the rest; both from accounts you own.
Poll the inboxes, classify replies, hand the warm ones to a human.
Sending infrastructure you own: a verified domain and enough inboxes for the daily limit. Charged once, reused on every run.
The receipt
About $9.65 for the whole run, plus $4.66 once for the sending inboxes. No subscription and no seats: each call is paid from a prepaid balance at the price shown before it runs.
| Capability | Calls | Cost |
|---|---|---|
| social.search$0.002 per result | 400 | $0.80 |
| social.comments$0.002 per result | 1,200 | $2.40 |
| social.profile$0.002 per match | 120 | $0.24 |
| company.enrich$0.0071 per match | 120 | $0.85 |
| contact.find$0.0187 per match | 84 | $1.57 |
| contact.verify$0.002 per verdict | 84 | $0.17 |
| web.scrape$0.0056 per page | 120 | $0.67 |
| social.posts$0.002 per result | 360 | $0.72 |
| email.send$0.0035 per email | 228 | $0.80 |
| linkedin.invite$0.0088 per invitation | 120 | $1.05 |
| linkedin.message$0.0088 per message | 36 | $0.32 |
| email.messages$0.002 per read | 23 | $0.05 |
| linkedin.messages$0.002 per read | 12 | $0.02 |
| One run | 2,907 | $9.65 |
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: Mine Reddit and LinkedIn for people complaining about their applicant tracking system, 400
npm install -g routergrowth && routergrowth setup
# one-time setup
routergrowth run -c email.domain -i '{"domain":"<domain>"}'
routergrowth run -c email.inbox -i '{}'
# 01 Find the conversations (social.search x 400)
routergrowth run -c social.search -i '{"platform":"<platform>","query":"<query>"}'
# 02 Read the threads (social.comments x 1200, social.profile x 120)
routergrowth run -c social.comments -i '{"url":"<url>"}'
routergrowth run -c social.profile -i '{"platform":"<platform>","handle":"<handle>"}'
# 03 Attach the companies (company.enrich x 120)
routergrowth run -c company.enrich -i '{"domain":"<domain>"}'
# 04 Get the contact (contact.find x 84, contact.verify x 84)
routergrowth run -c contact.find -i '{"first_name":"<first_name>","last_name":"<last_name>","company_domain":"<company_domain>"}'
routergrowth run -c contact.verify -i '{"email":"<email>"}'
# 05 Context for the first line (web.scrape x 120, social.posts x 360)
routergrowth run -c web.scrape -i '{"url":"<url>"}'
routergrowth run -c social.posts -i '{"platform":"<platform>","handle":"<handle>"}'
# 06 Run the outreach (email.send x 228, linkedin.invite x 120, linkedin.message x 36)
routergrowth run -c email.send -i '{"inbox_id":"<inbox_id>","to":"<to>"}'
routergrowth run -c linkedin.invite -i '{}'
routergrowth run -c linkedin.message -i '{"text":"<text>"}'
# 07 Read the replies (email.messages x 23, linkedin.messages x 12)
routergrowth run -c email.messages -i '{"inbox_id":"<inbox_id>"}'
routergrowth run -c linkedin.messages -i '{}'
FAQ
social.search returns public posts matching the words a buyer uses when the problem hurts. social.comments reads the threads under them and social.profile tells you who the author is and where they work.
The profile gives the person's name, bio links and often their company. company.enrich confirms the company and contact.find returns a work email for that person at that domain, charged only on a match.
A short, relevant message about something a person said in public is ordinary outreach, and it works because it is relevant. Keep it to people who posted in a professional context and follow the law that applies to your recipients.
Fewer than post. The workflows assume about three authors in ten are worth profiling, and an email is found for roughly seven in ten of those. The page shows the funnel for each one.
About $9.65 for one run of 2,907 calls, plus $4.66 once for the sending inboxes. 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/listening-ats-complaints.md · every workflow: /workflows.md
Paste the prompt into your agent. The first calls run on the $1 that comes with signup.