AdvancedUpdated Oct 18, 2026
Reps ask ChatGPT which account to work this week; Clay answers from live signals
Ops exposes Clay signals and a contact waterfall through MCP so reps get a ranked account list and verified contacts in chat.
RevStack prompt
Copy this into the model
Written by RevStack for this job. It is separate from the sourced result on this page.
Inputs
- Your territory or account list
- The signals Ops exposes
- How many accounts you can work this week
Using the Clay connector, rank the accounts in {{territory}} I should work this week based on these signals: {{signals}}. Return the top {{n}} with the signal that put each on the list, its date, and two contacts to engage. Do not include accounts with no signal in the last 30 days.Example output
An illustration of the shape of the answer. Not a reported result.
1. Account A: two new security hires in the last 14 days; pricing page visits. Contacts: CISO, Director of IAM. 2. Account B: opportunity moved to evaluation last week. Contacts: VP IT, Security Architect.
Several systems, or a change to how the team works.
The problem
SDRs with huge territories spent up to an hour per account deciding where to focus and finding contacts.
How AI solves it
Unify buying triggers in Clay (key page visits, new security hires, CRM stage changes).
Turn on Clay's MCP for reps inside ChatGPT or Claude.
Rep asks in plain language which account to work this week and gets a ranked answer with contacts.
Contact requests route to an Ops-built waterfall across many data providers.
Ops controls which workflows each rep sees and sets usage limits.
Tools that do this
Workflow
Step 1
Pick the 3 to 5 signals that predict a good week to call
Step 2
Build one Clay Function per rep request
Step 3
Pilot with a few reps and set credit limits