TrackIQ: Amazon DSP Audience and Creative Performance

get_dsp_performance takes a dimension argument that almost nothing uses: audience, creative, inventory, technology, geography. That is a whole data surface sitting unexamined in most accounts.

Which audience segments convert, which creatives carry them, and which supply sources waste impressions.

Part of Amazon AMC & DSP in the TrackIQ skills catalog.

Built as an Agent Skill. Runs in Claude Code, Claude web, Claude desktop and ChatGPT from the same folder.

---

Powered by the TrackIQ MCP

This skill reads your live Amazon account through the TrackIQ MCP — 16 tools connecting your AI assistant to Amazon data:

Sales & Traffic · Orders · Inventory · Returns · Sponsored Products · Sponsored Brands · Sponsored Display · Amazon DSP · AMC Cloud · Keywords · Search Terms · Targeting · Search Query Performance · Organic Rank · Best Seller Rank · Buy Box History · Brand Analytics · Export

Works with Claude, ChatGPT and Cursor. Get access →

---

What you get

Breaks Amazon DSP down by audience, creative, inventory source, technology and geography to show which segments actually recruit customers and which burn impressions. Ranks on new-to-brand purchases, detail page views and cost per action rather than ROAS, because DSP revenue is not returned at these dimensions. Use when the user asks about DSP performance, which audiences work, creative performance, DSP audiences, supply or inventory sources, where DSP budget is going, or why DSP is or is not working.

The rules that keep it honest

• sales is 0.0 and units is "0" on every row at these dimensions
• Never sum segments to a total
• Rank on new-to-brand purchases first
• Report cost per action, not efficiency ratios

The full list is in SKILL.md, and each one exists because getting it wrong produces a confident, wrong answer rather than an obvious error.

Requirements

• The TrackIQ MCP, for list_marketplaces and get_dsp_performance. - A DSP account with spend in the window. Not every account runs DSP. Check before promising a report. - Nothing else. No filesystem, no shell, no internet. - Without the MCP: works from a DSP report export broken out by the same dimensions.

---

Install

Claude Code

/plugin marketplace add TrackIQ-HQ/amazon-seller-skills
/plugin install trackiq-amazon-dsp-performance@trackiq

Claude web, desktop, mobile

1. Download the .zip from the latest release
2. Settings → Capabilities → Skills (code execution must be on)
3. Create skill → Upload a skill, choose the .zip
4. Toggle it on

ChatGPT

Same zip. Plugins → Skills → Create → Upload from your computer.

---

Setup

Answers live in account.md, copied from assets/account.example.md. Every TrackIQ skill reads the same file, so an account already set up for another TrackIQ report needs nothing added.

Delivery

Asked once and stored in account.md: in-chat (default), file, Slack, n8n or email. Anything leaving the chat confirms with you first and falls back to in-chat, with a note.

---

Customizing

File	What it controls
checks.md	the pre-send checks
method.md	the method and every threshold
pulls.md	the call sequence and its traps
report-template.html	the report shell

---

Contributing

python scripts/validate.py    # must exit 0 before any commit
python scripts/build.py       # writes dist/ zip + registry.json

Read AUTHORING.md before proposing changes.

License

MIT. See LICENSE.