# AI Intelligence Overview

Analyze conversion, quality, and coaching signals across pathways. Read the AI Intelligence guide in Hello Alex documentation.

Canonical HTML: https://docs.helloalex.ai/article/ai-intelligence-overview

Path: [Documentation](/) / [AI Intelligence](/ai-intelligence)

## Overview

AI Intelligence helps teams improve outcomes because it translates call behavior into prioritized coaching and optimization signals.

## How it works

- Aggregated trends surface shifts in quality, conversion, and transfer performance over time.
- Pathway- and call-level drill-downs connect signal changes to concrete examples, enabling targeted edits rather than broad guesswork.
- Generate Analysis runs a fresh AI pass over recent calls, while Auto Generate Analysis queues analysis for each newly completed call after you confirm billing.
- From Tags turns Call History transcript tags into pathway fix suggestions you can review and apply from the same drawer used on Insights.

## Use cases

- Weekly quality coaching
- Conversion optimization

## Tips

- Good to know: insight review is most effective when tied to a recurring optimization cycle after launches.

## Overview

Overview consolidates top signals so teams can prioritize improvement work quickly.

Open the highest-impact recommendation and trace it to pathway-level contributors.

- Headline recommendations summarize where outcomes shifted, reducing time spent on manual report assembly.
- Trend movement highlights whether recent launches or pathway edits produced meaningful changes.
- **Avg Score:** Aggregate quality signal.
- Good to know: reviewing overview after each major campaign provides faster feedback loops.
- Executive snapshot reporting
- Weekly optimization planning

## Pathway Insights

Pathway insights tie performance changes to specific conversation designs.

Implement one suggested improvement and compare week-over-week metric movement.

- Per-pathway metrics show booking or conversion movement over comparable periods.
- Suggested copy or structure changes provide practical hypotheses for improving weak steps.
- **Booking rate:** Share of calls reaching goal.
- Good to know: isolated A/B edits on opening lines often reveal high-leverage gains quickly.
- Pathway iteration cycles
- Goal-completion improvement

## Call Quality

Quality analysis exposes where conversations degrade and where escalation logic needs refinement.

Review low-scoring examples and apply targeted pathway or routing updates.

- Dimension-level scoring identifies weak areas such as clarity, transfer accuracy, or completion consistency.
- Linked examples in call history make it easier to validate whether observed issues are systemic or isolated.
- **Transfer accuracy:** Warm transfer success rate.
- Good to know: pairing quality findings with pathway edits shortens time-to-improvement.
- Transfer reliability optimization
- Quality postmortem analysis

## From Tags

From Tags converts tagged call moments into pathway improvement suggestions.

Tag a recent problem call, open From Tags, and generate a suggestion for the highest-impact tag.

- Tag calls in Call History using AI Error, Needs Correction, Unclear Audio, or Great Response markers.
- Non–Great Response tags appear in the From Tags queue where you can click Generate suggestion for an AI pathway fix.
- Review and apply accepted suggestions from the same apply drawer used on Insights, then track outcomes under Applied Changes.
- **Generate suggestion:** Requests an AI pathway fix for a tagged moment.
- **Auto Generate Analysis:** When enabled, newly completed calls are analyzed automatically after a billing confirmation.
- Good to know: tag at least a handful of calls before relying on From Tags suggestions for pathway edits.
- Targeted pathway fixes from real call examples
- Coaching loops tied to transcript tags

## Next step

Pick one underperforming pathway metric and run a focused edit-and-retest loop.

## Related documentation

- [More AI Intelligence guides](/ai-intelligence)