An Outbound Agency Has Grok Bot Build Clay Workflows and Review Every Campaign Daily
An agency lets helpers build email lists, then checks each day which ones are weak.
What happened Eric Nowoslawski's agency, Growth Engine X, has Grok Bot build Clay workflows from chat and review live campaigns daily. For client ClickUp, campaigns went from 10 positive responses a day to about 40, per Clay's blog.
Who this is about: B2B outbound agency (Growth Engine X; Eric Nowoslawski)
What it takes
- Cost
- From $30/monthGrok Bot: Premium is $30 a month, Super $100 and Ultra $200, each billed monthly. Grok Bot is included with an eligible Premium, Super, Ultra, Teams or Enterprise plan.You also need a Clay plan with CLI access (Admin or Editor role) and a sending tool; the blog gives no costs.Price checked Oct 24, 2026 on x.ai. Prices change; check before you buy.
- Setup time
- A few daysOur estimate from the source, for one business.
- Skill needed
- Some tech skillsComfortable installing an app, pasting an API key or running a command.
Explain it simply
An agency sends email to people who did not ask for it, and each job goes to its own helper. The helpers build lead lists and email steps in Clay, a data tool, from a short note. Each day a check looks for campaigns that miss their goal. It suggests a new list or new words, inside limits the agency sets.
What was done
He runs one Grok Bot per job or client: a copywriting bot per campaign, a list-building bot, a daily deliverability bot that drops the worst 30% of sending domains, and client bots that check at-risk accounts at noon and 6 p.m.
With Clay's agent plugin installed, Grok Bot has the Clay CLI and builds workflows in Clay from a chat or voice description; he then checks them in Clay.
For two Polyshares campaigns he gave a bot the audiences, copy and guardrails (validated emails only, unique leads, drafts until he says go live). It pulled 400 firms per campaign and built the pipeline into SmartLead.
A daily routine, run through the Grok Build CLI on the Bot's computer, compares each live campaign with its positive-response target and suggests list or copy changes within allowed messaging and filters.
How it went
For a PR client, the loop found that naming an industry publication for each prospect got 40% more responses than naming the Wall Street Journal and TechCrunch. For ClickUp, campaigns went from 10 positive responses a day to about 40. Clay doesn't give the time period or volumes.
- Promotional: it's Clay's own blog, by its director of growth. Eric is a former early Clay employee whose agency sells outbound, and he is an affiliate of Polyshares, one of the campaigns shown.
- The ClickUp and PR-client figures are reported secondhand, with no baseline period or volume.
- His rules: define the ideal customer profile (ICP) and the goal (meetings, not replies), prepare all the data up front, and keep humans writing the email templates.
Prompt
Every morning, review the live campaigns for [client].
Allowed messaging: [the hand-written templates and the one slot you may vary].
Allowed list filters: [ICP definition and filters].
Goal: [positive replies / booked meetings] per day, target [number].
For any campaign below target, suggest one change to the list or one change to the variable slot, with your reasoning and the data behind it. Don't change anything live; put suggestions in [channel] for my approval.Adapted from the daily instruction paraphrased in Clay's blog.
Setup notes
- Define the ICP and the real goal before the loop runs; Eric says the bot otherwise softens emails to get replies instead of meetings.
- Enrich the data up front, in your own database or a Clay Audience, so the bot doesn't pull or pay for data mid-run.
- Keep campaigns in draft until you say go live, and only send to emails that pass validation.
Business fit
Who this helps, and how:
Your turn
Ideas to try on your own work. These are starters, not results anyone has measured.
- Which limit should a helper never cross when it builds a lead list?
- What would you want a daily check to flag before the email changes?
- Where do you want to read the steps before they go live?
Related
- Grok Bot + Clay: build and check GTM workflows through the Clay CLIStack
- Evidence grade DA Founder Automated Outbound and Meta Ads With Claude Code and Hermes: 100 Leads Contacted a DayLight evidence, using Claude Code, Hermes Agent and Meta Ads CLI
- Evidence grade BA Consultant Turns a Company Name Into a Scored Lead in Under 20 MinutesStrong evidence, using Gemini, NotebookLM and Google Sheets
- Evidence grade CA Revenue Team That Had Never Shipped Code Built Its Own Call Review Tools in Claude CodeVendor or secondhand, using Claude Code, n8n and Clay
- AI for sales prospecting examplesDepartment hub: AI for sales, 34 examples
Similar tools
Other tools for this job, and how they differ. Chosen by our editors, never paid for.
- Claude Code
Clay's docs list Claude Code as a plugin host; same CLI, but it runs where you run it rather than on an always-on cloud computer.
Source: Clay
- Codex
OpenAI's coding agent, also listed by Clay as a host for its agent plugin and CLI.
- n8n
A fixed, scheduled workflow for the daily campaign check; predictable, but without an agent reasoning about changes.
“Vendor or secondhand” (grade C) means: A tool maker's case study, a promotional demo, or someone else's write-up. How we grade. Spotted a mistake? Suggest a correction.