Send each event attendee a personal agenda built from your account data
Before a conference, an agent reads what you already know about each registered attendee and drafts a short plan of which sessions to attend and who to meet. The account owner edits it and sends a private link instead of the full event PDF.
Paste this into ChatGPT, Claude or your agent tool. See the prompt
- Content and Collateral
- intermediate
- AE
- Sales leader
- CS
- Category
- Content and Collateral
- Difficulty
- Intermediate
- Est. cost
- Not stated by source
Problem
Every attendee gets the same full agenda, so the event does little to move a specific account forward. Reps rarely have time to tailor a plan for each customer or prospect who registered.
Setup
Clay holds the attendee list tied to each account and owner, along with signals the company already stores, such as product usage, call snippets and recent public posts. A Clay Agent loaded with the full conference agenda drafts a short plan for each person: the talks and roundtables that fit their goals and role. A second pass across all attendees suggests peers with a similar problem worth meeting. The data is sent to a content management system that generates a private page per person, which the account owner can edit and publish, and the owner is notified to send it to the customer before the event.
Prompt
Starter prompt by The GTM Agent Weekly, written from the setup described on this page. It is not the source's own prompt; see the source link below for their write-up.
# ROLE
You build a short, personal event agenda for each attendee from our accounts, so the account
owner can send it before the conference.
# INPUTS
- {{AGENDA}}: the full conference agenda (talks, roundtables, times, speakers, descriptions).
- {{ATTENDEE}}: name, title, company, account owner.
- {{SIGNALS}}: what we already know: product usage, recent call snippets, recent public posts.
- {{ALL_ATTENDEES}}: the list of other attendees with short profiles (for the peer pass).
# TASK
1. From {{SIGNALS}}, infer the attendee's likely goals in two short bullets. Only use what is there.
2. Pick three to five sessions from {{AGENDA}} that fit those goals and their role, with one
line each on why.
3. Peer pass: suggest up to two other attendees facing a similar problem, with a one-line reason.
4. Write a two-sentence intro note from the account owner.
# OUTPUT FORMAT
Hi {{FIRST_NAME}}, ...(intro)
YOUR SESSIONS: time, title, why it fits.
PEOPLE WORTH MEETING: name, company, why.
# GUARDRAILS
- Never mention private data in the text (usage numbers, call quotes). Use it only to choose.
- Only suggest sessions that exist in {{AGENDA}}, with correct times.
- The account owner reviews and edits before anything is published or sent. Replace every {{PLACEHOLDER}} before running. Works with any capable LLM and runner.
Tools
Steps
- Start from the registered attendee list, tied to each account and its owner.
- Pull the signals you already hold on each person: product usage, call snippets and recent public posts.
- Give a Clay Agent the full conference agenda and have it draft a short plan per person, not the whole schedule.
- Run a pass across all attendees to suggest who else shares a similar problem and should meet.
- Send the data to a content management system that builds a private page per person that the account owner can edit.
- Notify the owner to review, publish and send the link before the event.
Results
Described as running for attendees of Clay's own conference; no outcome data shared.
What goes wrong
Gotchas
- A wrong LinkedIn match or a board seat listed instead of the person's day job will poison the draft, so let owners overwrite before publishing.
- Companies that share a domain can get merged; keep each plan tied to one account.
- Do not suggest peer introductions that the stored signals do not support.
- A peer intro shares one attendee's interests with another, so check what each person would be comfortable having mentioned. (editor note)
Get 5 like this every week, free.
Real agent setups with sources and failure modes, in one short weekly email.
Free. One email a week. Unsubscribe anytime.
Source and credit
newsletter.gtmengineering.ai ›
Alex Lindahl, The GTM Engineering Newsletter · newsletter
Source dated 2026-10-09 · last reviewed 2026-10-12. Summarized in our own words; see the source for details.