NowGTM at Claimlane
Latest postLatest from the blogQualify Demo Requests in Clay with GPT-6 Luna
Every inbound demo request looks the same in your inbox — name, email, maybe a company name. Reps waste time chasing requests that were never a fit. This build routes each submission into Clay, enriches it automatically, and asks GPT-6 Luna to score ICP fit so the best leads are already highlighted when a rep checks in.
Create a Clay account and open a blank table
Clay is a spreadsheet-like tool where each row is a lead and each column is a piece of data about that lead. Go to clay.com, sign up for free, and click + Create New on your home page. Choose Blank table to start empty.
Add a webhook source so form submissions land here
A webhook is a link your form can POST data to — think of it like a letterbox Clay gives you. When someone fills in your demo form, Clay receives the details instantly and adds a new row. In your blank table, click + Add at the bottom, search for Webhooks, and click Monitor webhook. Clay displays a unique URL — copy it.
Connect your demo request form to the webhook
Open wherever your demo form lives — a Typeform, a HubSpot form, Tally, or even a plain HTML form. Find its Notifications or Integrations settings and paste the Clay webhook URL. Send a test submission so Clay receives one sample row. You will see it appear in your table within seconds.
Enrich company data with Claygent
Claygent is Clay's built-in AI web agent. You give it a prompt and it browses the web for each row, returning facts about the company. Click the blue Add enrichment button, search for Claygent, and select it. Write a prompt that uses the company column from your form. Clay will run it row-by-row automatically.
Add an AI scoring column with GPT-6 Luna
Now add a second AI column that reads the enriched data and outputs an ICP score. An ICP (Ideal Customer Profile) is a description of the type of company that tends to become your best customer. Click Add enrichment again, select AI from the sidebar, then choose Use AI. In the model dropdown, pick GPT-6 Luna — it is the cheapest GPT-6 model and OpenAI designed it specifically for high-volume extraction tasks like this. Write a scoring prompt.
OpenAI · GPT-6 Luna · ~$0.10 / 1M tokens
Sort and filter so reps see Hot leads first
Clay tables work like spreadsheets. Click the column header on your tier column and choose Filter to show only rows where tier equals 'Hot'. You can also click Sort on the score column to sort highest first. Now your table is a live, auto-scored queue — every new form submission flows in and gets a score without any manual work.
What you get
You end up with a live Clay table where every new demo request is automatically enriched with company data and scored for ICP fit by GPT-6 Luna. Reps open one tab instead of triaging raw inbox emails, and the hottest leads are already at the top. Rough effort: ~90 minutes, Clay free plan (includes trial credits), GPT-6 Luna API costs fractions of a cent per lead at $0.10/M input tokens.