# Build privacy-aware audience segments

First-party customer data often lacks a clear, reusable audience plan. This skill turns approved CRM or GA4 exports into named seeds, exclusions, and funnel-stage targeting maps.

## Install

```bash
npx skillstore add aaron-he-zhu/audience-segment-builder
```

## Metadata

- Status: approved
- Slug: aaron-he-zhu-audience-segment-builder
- Version: 19.0.0
- Author version: 19.0.0
- Skillstore revision: r2
- Version status: valid
- Tree hash: f180309da78396613ef4b54b12a14d1e67b9422a632b763a8c902ccd52e51ae6
- Author: aaron-he-zhu
- GitHub username: aaron-he-zhu
- License: Apache-2.0
- Repository: https://github.com/aaron-he-zhu/aaron-marketing-skills/tree/main/ad/research/audience-segment-builder
- Ref: adcb3549b15782055d0beb6d29f113d20de08f92
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands, network, filesystem
- Quality score: 81
- Quality tier: silver
- Public page: https://skillstore.pages.dev/skills/aaron-he-zhu-audience-segment-builder
- Manifest: https://skillstore.pages.dev/api/skills/aaron-he-zhu-audience-segment-builder/manifest

## Capabilities

- Identifies available CRM or GA4 columns for segmentation.
- Defines named seed audiences tied to customer traits or behavior.
- Ranks high-value customer groups for value-based lookalike seeds.
- Creates exclusion and suppression rules for customers and recent purchasers.
- Maps cold, warm, hot, and customer audiences across ad platforms.
- Flags missing data as NEEDS\_INPUT instead of guessing.

## Use Cases

- Prepare a prospecting launch: Build high-value seeds and cold audience definitions before a new acquisition campaign.
- Reduce wasted retargeting spend: Define recent-purchaser and existing-customer exclusions with clear suppression windows.
- Standardize platform targeting: Translate GA4 or CRM signals into one funnel map for Google, Meta, and other platforms.

## Prompt Templates

### Beginner: segment a customer export

```
Build audience segments from my customer export. My goal is direct response. Identify useful columns, name each segment, and flag missing data.
```

### Intermediate: create value-based seeds

```
Use my CRM export to define value-based lookalike seed audiences. Rank groups by lifetime value, state the seed criteria, and create exclusions for current customers.
```

### Advanced: map a multi-platform funnel

```
Map my GA4 and CRM audiences to cold, warm, hot, and customer funnel stages for Google and Meta. Include retargeting windows, suppression rules, and NEEDS_INPUT flags.
```

### Advanced: review privacy-safe output

```
Create a privacy-aware audience segment handoff from this export. Use aggregate descriptions only, do not repeat raw PII, and list every source column used.
```

## Limitations

- It does not upload audiences to advertising platforms.
- It does not create campaigns, ad groups, or match types.
- Segment quality depends on the completeness of supplied first-party exports.
- It does not replace legal review of consent, privacy, or platform policies.

## Best Practices

- Use only approved first-party exports and confirm data-use permissions before segmentation.
- Tie every audience definition to a supplied column, value, or stated behavior.
- Use aggregate descriptions or hashes and keep raw PII out of the handoff.

## Anti Patterns

- Do not invent value tiers, retargeting windows, or fit signals that are absent from the export.
- Do not include raw email addresses, phone numbers, or customer rows in the output.
- Do not treat segment definitions as a substitute for campaign structure or platform policy review.

## Security Audit

- Audited at: 2026-07-26T09:57:12.51\+00:00
- Summary: All 31 static detections are false positives. Backticks are Markdown formatting, URLs are metadata links, and relative paths are documentation references or an explicitly confirmed local save location. The skill instructs users to treat exports as untrusted and avoid returning raw PII.

## Stats

- Views: 1
- Downloads: 5
- Favorites: 0
- Popularity score: 0