Performance

Documentation Index

Fetch the complete documentation index at: https://docs.agentweb.pro/llms.txt

Use this file to discover all available pages before exploring further.

Performance is where marketing results turn into decisions. Five sub-sections covering different data sources.

​

Research Reports

The Research Reports view aggregates audience research, market analysis, and Emma-generated insights. Reports generated in Agent Mode land here automatically. How to use:

• Search reports by title or keyword.
• Download or share to Slack.
• Mark favored reports with the heart icon. Emma uses these as positive examples for future generations.
• Open a report to edit inline.

​

Campaign Performance

Cross-channel campaign analytics. Spend, reach, conversions, ROI — by campaign, objective, and channel. How to use:

1. Select a time range and ad account.
2. Filter by campaign objective.
3. Compare campaigns across channels.
4. Identify winners to scale and losers to cut.

Watch-outs:

• A campaign’s right metric depends on its objective. Brand-awareness campaigns are not judged by CPL.
• Cross-channel ROI is hard to measure precisely. Don’t over-rotate on last-click attribution.

​

Paid Social

Meta and LinkedIn ad creatives, audiences, and active campaigns. How to use:

1. Review active and paused campaigns by account.
2. Drill into individual creatives — performance, audience targeting, spend.
3. Ask Emma for recommendations: “What should I cut from my Q2 campaigns? What’s working that I should scale?”

Watch-outs:

• Optimize based on the campaign’s stated objective.
• Give campaigns enough data before making changes. Pausing at 100 impressions is noise.
• Test creatives at low budget before scaling. A 20% lift at $50/day does not mean a 20$500/day.

​

Newsletter

Email newsletter drafts, audiences, and per-send performance. What’s tracked:

• Open and click-through rates per send.
• Subscriber engagement events.
• Per-issue content performance.

How to use:

1. Select a newsletter or campaign.
2. Review engagement metrics by send.
3. Compare subject lines and CTAs across issues.
4. Ask Emma to draft the next issue using the patterns that worked.

Watch-outs:

• Clicks matter more than opens. Apple Mail Privacy Protection inflates opens.
• Consistency beats frequency. Sending twice a week sporadically performs worse than every other week reliably.
• Subject lines and the first 2 sentences are where most readers decide whether to keep reading.

​

SEO / GEO

Search visibility plus presence in AI-generated answers (GEO = Generative Engine Optimization). What’s tracked:

• Keyword and topic ranking trends.
• Content performance by topic.
• Citations in AI-generated answers (ChatGPT, Perplexity, Google AI Mode).

How to use:

1. Monitor keyword and topic performance over time.
2. Identify pages gaining or losing visibility.
3. Check which content gets cited in AI answers.
4. Prioritize content updates and new topics based on the gaps.

Watch-outs:

• Focus on intent, not just keyword volume. A high-intent term beats a high-volume informational one for revenue.
• Update existing content before creating new pages. Refresh tends to outperform expansion.
• Structure content clearly. AI engines parse headings, lists, and tables more reliably than wall-of-text prose.

​

Engagement time metric

The engagement time shown on the SEO/GEO page is total engagement time across all visitors for the selected period, not per-visit average. It comes from Google Analytics 4’s userEngagementDuration metric. Example: if 100 users each spent 30 seconds on a page, the displayed time is approximately 50 minutes. To estimate per-visit engagement, divide the displayed time by the page’s session count for the same period.

​

Next steps

• Connect Meta so paid social data flows in automatically.
• Connect LinkedIn for organic post performance.
• CRM for lead-side metrics that pair with these channel metrics.