Houston AI

How this works

This overview summarizes system-wide performance over the selected date range.

Metrics are computed per lead exposure to avoid chatty leads skewing results.
Reply/negative rates are based on replies within the reply window.
Charts show trends; drill into templates for details.

Limits

No external benchmarks · Call outcomes may lag data ingestion

What to do next

Review the Template Scorecard to identify top opportunities.

Executive Summary

A fast readout of SMS performance and where attention is required.

Selected period

Executive takeaway: A small number of templates drive most friction—fix the highest-friction template first.

Three questions this page answers:

Is our messaging working?
Where are we losing leads?
What should we fix first?

Current State (Last Selected Period)

Reply rate

0.0%

Percent of exposed leads who replied

Negative reaction rate

0.0%

Percent of exposed leads with complaints/opt-outs

Templates needing attention

Templates flagged for improvement

What Stands Out

A small number of templates drive most friction.

No templates meet the exposure threshold yet.

Decision Overview

Each SMS template is evaluated using three signals: engagement, friction, and outcomes.

Rather than scoring individual messages in isolation, the system groups templates into behavioral clusters based on how recipients consistently respond over time.

Each cluster is assigned a verdict so the system reasons about patterns—not isolated events.

Verdict definitions

Keep — Patterns within expected performance ranges.
Improve — Recoverable issues where targeted changes should help.
Rewrite — Structural problems in intent/framing with consistent underperformance.

No verdict data for this range.

Where to Go Next