Automation

Score leads automatically from on-site behavior

A scenario prompt that turns a described behavioral lead-scoring workflow into a recipe with explicit point values, decay rules, and sales-alert thresholds.

Works with Claude / GPT240 uses 4

The prompt

automation-lead-scoring-site-behavior
A B2B marketing team describes their process like this: "I want to give a lead 10
points for visiting the pricing page, 5 points for downloading a whitepaper, and 20
points for visiting the pricing page three times in one week, and once someone hits
50 points I want sales alerted in Slack to reach out."

AVAILABLE TOOLS: [Site analytics/tracking pixel (trigger), custom scoring logic,
HubSpot (CRM field storage), Slack]

Turn this into an implementation-ready automation recipe, thinking through what the
plain description leaves ambiguous:
1. Trigger and scoring logic: specify exactly how "three pricing visits in one week"
   is detected without double-counting, since a naive implementation might award 10
   points per visit plus another 20 for the pattern, over-scoring a lead who visited
   exactly three times.
2. Conditions: decide whether points decay over time (a whitepaper download from four
   months ago probably shouldn't carry the same weight as one from yesterday), and
   specify the exact decay schedule if you recommend one.
3. Actions in order: write the score to a HubSpot custom property after each
   qualifying event, check the running total against the 50-point threshold, and fire
   the Slack alert only once per lead per threshold crossing, not on every subsequent
   point earned above 50.
4. Failure handling: if the tracking pixel fires twice for the same page view due to
   a browser back-button or a slow page load, how does the scoring logic dedupe
   that event before it inflates the score.
5. One edge case specific to this exact workflow: an existing customer's team member
   browsing the pricing page while comparison-shopping a competing product for their
   own research. Decide whether existing customers should be excluded from scoring
   entirely so sales doesn't get a false "hot lead" alert.

Replace the workflow description and available tools with your own.

Lead scoring almost always needs a decay rule and an existing-customer exclusion that never make it into the first draft of the workflow description. Since this usually means writing custom scoring logic beyond what a no-code trigger tool handles natively, CodeWP is worth a look for generating the scoring snippet itself.