Zapier + ChatGPT inbound lead summarizer (L1)

When a demo request comes in, Zapier sends the company website + form responses to ChatGPT, gets back a 5-line account brief (what they do, likely use case, 2 discovery questions), and posts it to Slack before the AE picks up the phone. Your first real AI-in-the-loop workflow.

WORKFLOW1Trigger on new demo reque…tZapier2Scrape the websiteZapier3Summarize with ChatGPTChatGPT4Post to the right AESlack
4 steps, in order, with the tool that owns each one.
Adoption ladderSix levels from Starter to Rebuilt. This item sits at level 1.L1 StarterOne tool, no workflow changeL2 AssistedAI drafts, humans approveL3 IntegratedWired into CRM and SlackL4 OrchestratedMulti-step, owned, measuredL5 AutonomousAgent runs, human auditsL6 RebuiltThe process itself changes
This playbook belongs at L1 Starter. Running it above your level is how pilots stall.
Measures of successdemo show rate; rep prep time per meeting; SQL conversionPROVE IT WORKEDdemo show raterep prep time per meetingSQL conversion

The steps

  1. 01

    Trigger on new demo request

    Tool: Zapier

    In Zapier, create a new Zap. Trigger: "New Form Submission" from HubSpot/Marketo/Webflow/Typeform (whichever your demo form lives in). Test the trigger so you can see the actual field payload. Owner: RevOps. Time: 15 minutes. Pitfall: triggering on every form (newsletter, contact-us), you'll burn OpenAI credits. Filter to "Form Name = Request a Demo". DoD: Zap shows a real test submission with name, email, company, website fields.

  2. 02

    Scrape the website

    Tool: Zapier

    Add a Zapier built-in action: "Formatter > Utilities > Extract URL" then a second "Webhooks > GET" against the company website. (Or use the Apify/Scrapingbee Zapier integration for cleaner HTML-to-text.) Grab the home page text and trim to 2000 characters. Owner: RevOps. Pitfall: scraping JS-heavy SPAs returns empty strings, fall back to the company's LinkedIn /about page if main site is empty. DoD: the step output contains real prose about the prospect's business.

  3. 03

    Summarize with ChatGPT

    Tool: ChatGPT

    Add "ChatGPT (OpenAI)" action in Zapier. Model: gpt-4o-mini (cheap + fast). Prompt: "You are a sales engineer. From this website text and form answers, produce: (1) one-sentence what-they-do, (2) most likely use case for our product [insert your product], (3) 2 specific discovery questions referencing their actual business, (4) one risk or red flag if visible. Be concrete, no generic SaaS talk." Pass website text + form fields as variables. Owner: RevOps. Cost: ~$0.001 per summary. Pitfall: forgetting to set max_tokens (300 is plenty), costs balloon on long sites. DoD: test output looks like a real human brief, not corporate boilerplate.

  4. 04

    Post to the right AE

    Tool: Slack

    Final Zapier action: Slack > Send Direct Message to the assigned AE (look up by round-robin or territory in a HubSpot lookup step). Include: prospect name, company, meeting time, the AI brief, and a link to the deal record. Bonus: also post in #inbound-firehose for the manager to skim. Owner: RevOps. Pitfall: posting in a public channel only, AEs miss it. DM is non-negotiable. DoD: AE confirms they received the brief and used at least one discovery question on the call.

Tools in this playbook

Next playbooks

Unfamiliar terms are defined in the AI and Revenue Dictionary. Related frameworks live in the framework library.

Share this playbook

Posting to Instagram or TikTok? Copy the link, it carries the title, summary and share image.

Arrives weekly by email. Free. Unsubscribe anytime. By subscribing you agree to our Privacy policy and Terms. We never sell or share the list.