Put an AI agent on your highest-intent page to qualify and book buyers, plus a personalised self-serve page a person approves
SaaStr's account of a year of inbound AI: an agent on the sponsor page qualifies visitors and books meetings, and a second path sends a company-specific page, watches the visit, and drafts a pitch that a person approves and sends.
Evidence: The author reports this. We have not checked it beyond reading the source.
The business problem
High-value inbound buyers land on a pricing or sponsor page, fill in a form and wait for a rep. Some will talk to a chat agent, others only want to read, and a small team cannot follow up on every visit quickly or personally.
What was tried
1. Place the on-site agent (built on Qualified) on the page where buyers already evaluate a purchase, not the homepage. 2. Have it qualify in real time: why they want to buy, budget, goal, competitors considered and the results they want, then book the meeting itself and pass the conversation to the human so the call starts from what was said. 3. Offer a self-serve option next to it for people who will not chat. 4. Replace the static PDF with a tokenised, company-specific page on your own site, add heat mapping (Microsoft Clarity) and wait about 10 minutes after a visit starts before acting. 5. Check first-party signals first (site visits, email, ads, events, newsletter), then competitors, then third-party data, and route the lead with the heat map to the owner by email, Slack, a dashboard queue and Salesforce. 6. The agent drafts a pitch using only public or anonymised customer results; a person reviews and approves it and it is sent from them, and the same link updates so a returning visitor sees the tailored version. The backend runs on Replit, and a custom booking tool replaced Calendly.
What was reported (positive)
SaaStr reports about 17,000 conversations and about 600 meetings booked in 12 months for its annual event's sponsorships, and says the inbound agent plus the self-serve path raised new business by 60 percent. Three people run the system. Quality dropped for a few weeks when the backend grew too large, fixed by splitting it up. The 12-month funnel table did not load, so conversion rates could not be checked.
Limitations
A media and events company with a team, selling sponsorships worth tens of thousands of dollars, not a solo founder; the author says it suits tech-savvy buyers and may not work for others. All figures are self-reported, many conversations were not serious, and other published case studies give different numbers for SaaStr's inbound. SaaStr sells sponsorships and names Replit as a top sponsor; no deal with Qualified is disclosed. No model or prompt details, setup costs or publish date are given; one lead example is a single anecdote.
What you need
A clear high-intent page with enough traffic, a CRM, and someone who can review and send drafted pitches quickly.
Sources
Tools in this workflow
- Salesforce: CRMTry Salesforce
- Replit: Backend, dashboard and CRM write-backTry Replit
- Qualified: On-site AI agent that qualifies and booksTry Qualified
- Microsoft Clarity: Heat maps of the self-serve pageTry Microsoft Clarity
Get new workflows like this by email
A short selection of the most useful new workflows, including the ones that did not work. Free, and you can unsubscribe at any time.
Related workflows
Test a Claude-written sales page headline against your own (copywriter reports a win, and a loss on subject lines)
A freelance-copywriting newsletter author gave Claude Opus 4.6 basic context about his offer, put its headline up against his own on his sales page, and reports about 50% more conversions. He says AI lost to him on email subject lines.
Make OG images and app store screenshots with an AI image editor for the art and code for exact sizes and text
A side-project developer splits launch graphics into two jobs: an AI image editor makes backgrounds and mascots with no text, and a small script composites real screenshots, renders headlines and checks every platform's size rules.
Turn client meeting notes into a branded A4 proposal PDF with a Claude Code skill and Puppeteer
An agency owner shares a Claude Code skill that parses meeting notes, fills a multi-page HTML proposal, checks page heights and renders a PDF with Puppeteer. Its repository supplies the template and rendering setup.
Turn a messy website brief into a Claude-assisted proposal while setting prices and timelines yourself
An agency worker shares a five-stage proposal process: clarify the brief, define phased scope, supply a rate card, draft in the agency's voice and review as a skeptical client.