# Attify for support leaders

> For heads of support running an AI agent: the share of its resolutions that held for 7 days with no handoff, why the rest failed, and what each real resolution cost, every month.

Published 2026-11-12 · Reviewed 2026-11-12. Canonical: https://attify.app/for/support-leaders

Attify tells you, each month, what share of the conversations your AI agent counted as resolved actually held (no repeat contact within 7 days and no handoff, and, in a judged audit, confirmed by a judge on a sample), why the rest failed, and what each real resolution cost. It reads your own help desk, so the number does not depend on the vendor's dashboard.

## The questions it answers

| You need to know | Where |
|---|---|
| Is the vendor's resolution rate real? | The audited rate next to the claimed rate, with the gap in points |
| Where does the AI fail? | Why claims failed: repeat contact, handoff, not resolved, by contact reason |
| Is it getting better? | Month on month, with intervals, so a move inside the noise reads "no clear change" |
| What does it cost per real resolution? | Cost per audited resolution against the price per claimed one |
| Do we renew? | A signed renewal pack per agent and month (Team plan) |
| How do two agents compare? | Several agents of one job type side by side (Business plan) |

Improvement suggestions, with redacted examples of failed conversations, are shown to you about your own agents. Attify never shows them to a vendor.

## How it works

1. **Free check first.** Upload an export at [/check/support-resolution](https://attify.app/check/support-resolution) for a rules-tier result on your own data, no account needed.
2. **Connect your help desk.** Intercom, Zendesk, HubSpot, Salesforce, Gorgias or Freshdesk, read-only, with your own authorization.
3. **Get the month's verdict** by email on the 1st and on the dashboard: one sentence, three numbers, the reasons.

The rules are those of the [Attify Audited Resolution Standard](https://attify.app/standards/support-resolution), published with its judge prompt word for word.

## What to read first

- [How to measure your AI support agent's real resolution rate](https://attify.app/guides/measure-ai-support-agent-resolution-rate)
- [How to audit conversations your AI marked resolved](https://attify.app/guides/audit-ai-resolved-conversations)
- [How support AI vendors define and bill a resolution](https://attify.app/compare/support-ai-agent-outcome-definitions)
- [Resolution vs. containment vs. deflection](https://attify.app/compare/resolution-vs-containment-vs-deflection)

Plans, the free plan for buyers who contribute anonymous data, and the 14-day trial: [for buyers](https://attify.app/buyers).

## Questions

### Which help desks does Attify read?

Intercom, Zendesk, HubSpot, Salesforce Service Cloud, Gorgias and Freshdesk, plus voice agents on Vapi, Retell AI, ElevenLabs, Bland AI and Voiceflow, and any agent through the Ingest API.

### Does Attify see our customers' messages?

It reads them to check outcomes, redacts personal data before any text goes to the judge or into an example, and deletes text 30 days after the conversation's last update by default. Scores and counts are kept. Vendors never see your conversations.

### Will this replace our QA tool?

No. QA scores individual conversations against your rubric. Attify measures how often claimed resolutions held, under a published rule, and why the rest failed.

## Sources

- Attify: [Attify Audited Resolution Standard v0.11](https://attify.app/standards/support-resolution), read 2026-11-09
