AI Stack for B2B Retargeting

Salesforce lost-opp data + pixel audiences → Claude → Custom segments + retargeting copy in Notion
Matt Danese
Senior Demand Generation Manager. These stacks are built and used in production — not generated for a listicle.

Most retargeting audiences are built from pixel data alone — everyone who visited a page, regardless of whether they were ever a real prospect. This stack flips that. Pull your Salesforce lost and stalled opportunity data, cross-reference it against your ad platform pixel audiences, and let Claude identify which segments are worth retargeting and what message angle matches each loss pattern. The output is a segmented audience list and ready-to-use ad copy variants — not a generic retargeting campaign aimed at cold visitors who bounced in 12 seconds.

The Stack

Input
Salesforce lost/stalled opp data Ad platform pixel audiences
AI
Claude
Output
Custom audience segments Retargeting ad copy in Notion

The Prompt

This stack is built around the Bad Lead Pattern Detection Prompt. Here's the abbreviated version — the full prompt with all variables and usage notes is on its own page.

Claude Prompt — Abbreviated
You are a B2B demand gen analyst reviewing lost and stalled deal data.

Analyze the Salesforce opportunity data below and identify the patterns that should
inform retargeting strategy. For each loss reason cluster, recommend:
1. Whether this segment is worth retargeting (and why)
2. The message angle that addresses their specific objection
3. The ad format and channel best suited to re-engage them

Focus on segments with 3+ similar loss reasons — outliers are noise.
[ ... continued — see full prompt ]

The Workflow

  1. Export lost and stalled opp data from Salesforce

    Pull the last 90 days of closed-lost and stalled opportunities. Include: company name, deal stage at loss, loss reason, ICP attributes (company size, industry, job title), and days in pipeline. Export as CSV.

  2. Export active pixel audiences from your ad platforms

    Pull current audience lists from LinkedIn, Google, and Meta — which audiences are active, their current sizes, and what URL behavior or engagement signals they're based on. Note any overlap with your Salesforce export.

  3. Paste both exports into the prompt

    Give Claude the Salesforce opp data and your pixel audience overview together. Claude will cross-reference deal-loss patterns against audience composition and identify which segments are actually worth retargeting versus which are waste.

  4. Review segmentation and copy recommendations

    Claude flags segments by loss pattern — budget objection, timing, competitor loss, disqualified — and maps each to a specific retargeting message angle. Review the segmentation logic before building audiences.

  5. Build audiences and add copy to Notion

    Create new audience segments in your ad platform based on Claude's recommendations. Paste the generated copy variants into Notion for creative review and trafficking into your campaigns.

What This Replaces

Related Stacks

New stacks drop weekly.

Each one includes the tools, the Claude prompt, and the workflow logic. Free — built for in-house B2B demand gen managers.

← Back to Stack Library

© 2026 The Demand Engineer · thedemandengineer.com