Ask or search…

PLS: Spotting decision makers in your free trial signups with AI


A lead has just signed up to your product's free trial. but you're not sure if it's a decision maker worth closing?
This a use case from Cargo's AI community to solve this challenge using Cargo's native GPT-4 integration. The goal is to make sure the AI doesn't make too many mistakes using a good chain of thoughts.

Why this matters

Boost Sales Efficiency: Focus on prospects who have the authority to make purchase decisions
Strategic Triangulation: Identify key players in your prospect's team for building social proof ahead of the decision-making process
Customise Engagement: Tailor your approach based on the type of decision-maker (e.g., founder vs team manager)

Phase 1: Data Preparation:

  1. 1.
    Ensure you're capturing a field such as job title in your signup. It's beneficial to include team size
  2. 2.
    Make sure you're filtering out non-professional emails. Otherwise use a recipe from the Cargo community to filter these out for you:

Phase 2: Pit the AI vs a REGEX function

  • Step 1: Set up a backup classifier using REGEX (reach out to [email protected] the full prompt)
  • Step 3: Introduce an 'adjudicator' GPT classifier capable of revising the REGEX output (reach out to [email protected] the full prompt)

Phase 3: Refining results

  • The goal: Since GPT is prone to error every now and then, you'd want to bias the output slightly towards false positives. Remember, a minute spent on a mistakenly classified decision-maker is less costly than missing out on a potential lead
  • Step 5: Add a verification step comparing GPT vs REGEX
  • Step 6: For potential false negatives, consult the 'adjudicator' node
IF there's strong evidence, consider overruling the REGEX in favor of GPT's suggestion ELSE IF there's a consensus or risk of false positives, proceed with that output

Phase 4: Routing Decision Makers:

  • Found a decision-maker lead? Notify your sales rep via Slack and integrate with your CRM and outreach tools
  • Otherwise, consider enriching the lead to identify a stakeholder before integrating with your tools


Test this process initially, evaluate its performance, and once satisfied, roll it out in production
Need a template? I've been using one with Cargo clients in prod. Reply to this email for the full prompt, code, and template