# Socialseal Performance Readout

> Use this skill when the user asks how posted content, a campaign, a creator programme, or a content period performed in SocialSeal; what changed after publishing; or what the latest results mean. Produce a scoped readout with metrics, examples, interpretation, caveats, and next actions without claiming unsupported causality.

- Skill: `opensealai/socialseal-performance-readout` (Agent Skill, multi-file: 5 files)
- Install (CLI): `npx skillmds@latest add opensealai/socialseal-performance-readout`
- Raw SKILL.md: https://api.skillmd.com/api/skills/opensealai/socialseal-performance-readout/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Marketing & Growth
- License: MIT
- Author: OpenSealAI (https://skillmd.com/u/opensealai)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/opensealai/socialseal-performance-readout

---


# SocialSeal Performance Readout

## Overview

A SocialSeal performance readout explains what changed in discoverability and what to do next. It should connect posted content and tracked keywords without pretending SocialSeal data is a complete view of every platform interaction.

## Inputs

- current and comparison SocialSeal exports
- list of posted content during the period
- tracking group IDs and keyword/topic scope
- previous objectives or content plan
- optional platform-native metrics

## Data Access

```bash
npx -y @socialseal/cli data export-group-evidence \
  --group-id <group-id> \
  --workspace-id <workspace-id> \
  --out ./exports/group-evidence-current.csv
```

```bash
npx -y @socialseal/cli data export-search-results \
  --group-ids <group-id> \
  --workspace-id <workspace-id> \
  --date-from <iso> \
  --date-to <iso> \
  --out ./exports/search-results-current.csv
```

MCP-first: use `socialseal_export_report` (`reportType: "search_results_enriched"`) or `socialseal_export_tracking_data`; there is no group-evidence MCP tool. See `references/mcp-and-cli-usage.md`.

Attribution: cite movement in human-readable terms, the `"keyword" [market, platform]`, the video title/URL, `@author_handle`, and where it ranked; keep `video_uid`/`search_result_id` as an internal traceability note. Use `first_seen_at`/`last_seen_at` for resurfacing language and never infer platform age when `published_at` is blank. See `references/socialseal-data-contract.md`.

Evidence tiers: movement in discoverability/coverage/SOV is a hard observation, not a noisy estimate; attribute day-to-day change to a shifting search reality, not measurement error. But the figures are scoped to high-ranking videos for the tracked queries, so do not generalize past that scope. See `references/evidence-and-confidence.md`.

## Workflow

1. **State scope.** Platform, market, language, tracking group, keyword set, date range, and comparison period.
2. **Validate data.** Check rows, columns, date range, missing identifiers, and whether the export is social or Google AI evidence.
3. **Calculate metrics.** Keyword coverage, discoverability by topic, SOV where qualified attention exists, owned/creator split, competitor/entity changes.
4. **Connect to content.** Map posted content to tracked topics carefully. Use “consistent with” unless causality is established.
5. **Explain examples.** Include examples that show why a metric moved or why a gap remains.
6. **Write next actions.** Continue, adjust, add, pause, or investigate.

## Output

- scope and data note
- metric snapshot
- changes vs previous period
- examples and interpretation
- caveats
- next actions

Use `templates/measurement-readout-template.md` when a simple report format is needed.

## Do / Don't

Do:

- show denominators and comparison periods
- separate owned and creator content
- call out missing data clearly
- make recommendations specific enough to brief

Don't:

- lead with raw row counts or internal IDs
- claim causality from timing alone
- treat a hard measurement's day-to-day variation as noise
- generalize a scoped statistic to the whole platform/market
- hide export freshness or scope
- overload management audiences with every table

## Troubleshooting

- If IDs are missing, use group evidence or refresh enriched search results.
- If current and previous groups differ, state that comparison is directional only.
- If platform-native metrics conflict with SocialSeal signals, explain that they measure different surfaces.

## Verification Checklist

- [ ] Scope and data freshness are stated.
- [ ] Metrics use percentages/denominators where relevant.
- [ ] Examples support interpretation.
- [ ] Recommendations are concrete.
- [ ] Caveats are visible.