Were the resolutions you paid for real?
Drop your support export and see which billed AI resolutions failed a stricter test. Your file never leaves your browser.
Drop your export here
CSV or JSON from Intercom, Zendesk, Ada or Gorgias. Other helpdesks work with a column mapper.
Other samples: ,
Read in this tab. Nothing is uploaded, no account needed.
This is what you get back.
A real audit of the built-in Intercom sample: 774 AI conversations from Jul 9, 2026 to Sep 30, 2026, run by the same code as the tool above.
22% of billed resolutions failed the strict test.
- Billed as resolved
- 451
- Really resolved
- 352
- At $0.99 each
- $98.01
Why they failed
- Same customer came back53
- A human replied after the AI17
- Reopened20
- Negative CSAT17
Examples from the dispute file
- Phantomic_100020Repeat contactThe same customer opened conversation ic_100080 5.4 days later.
- Phantomic_100037Human replyA human replied after the AI in this conversation.
- Phantomic_100017ReopenedReopened 1 time.
- Phantomic_100003Negative CSATSatisfaction 0.25 on a 0 to 1 scale, below the 0.4 threshold.
Every flagged conversation lands in a dispute CSV with its reason, plus an email draft.
Three steps, about a minute.
Export
Pull a conversation export from your helpdesk. Intercom, Zendesk, Ada and Gorgias are recognised from the headers.
How each vendor exports and billsCheck
Each billed resolution is tested against five failure signals inside a window you choose, seven days by default.
Read the StandardDispute
Download the flagged conversations with a reason for each, plus an email draft asking for a credit review.
Estimate what it is worth
Five signs a billed resolution was not real.
A conversation passes only when none of them appear inside the window.
The customer came back
The same customer opened another conversation inside the window. The evidence names it and the gap in days.
A human stepped in
A teammate replied after the AI. The Standard counts it even where the vendor does not.
It was reopened
The reopen count is above zero, or the timestamps show it reopened inside the window.
The rating was low
Satisfaction below 0.4 on a 0 to 1 scale. On a 1 to 5 scale, a 2 counts as 0.25.
The vendor escalated it
The vendor's own state says escalated, handed off or negative feedback.
Why you can trust the numbers
Your file stays on your device
Parsing and scoring run in your browser. Open the network tab during an audit: no request is made.
Open source
MIT-licensed code you can read, run and fork.
Open definition
The Strict Resolution Standard is public and versioned, licensed CC BY 4.0.
Questions people ask first
What does Resolvara do?
It reads an export of your support conversations, finds the ones where an AI agent took part, and applies the Strict Resolution Standard to them. It shows how many of the conversations that could be judged your vendor billed as resolutions, how many of all those show no sign of failure, and which billed ones do.
Does my data leave my browser?
No. The file is read and analysed in your browser, in a Web Worker. The audit makes no network request. Your file and results are only ever sent if you click "Contribute anonymously", and then the exact JSON is shown to you first: aggregate rates only, no ids, no text and no customer data. Separately, when you leave the page it sends anonymous usage counts (which steps you reached, never file content). See the privacy page for how to confirm this in your browser's developer tools.
How accurate is it?
It is as accurate as your export and the signals allow. Four of the five fail signals are direct readings of the data (vendor escalation state, a human reply, a reopen count, a low rating). Repeat contact is a proxy: the same customer wrote again within the window, which usually means the first answer did not solve the problem but sometimes means a new problem. For that reason every flag carries its evidence and you can change the window, turn on the same-topic filter, and read the flagged conversations yourself.
Is a "phantom resolution" the same as fraud or an invoice error?
No. A phantom resolution is a conversation your vendor billed that shows at least one fail signal under the Standard. Vendors publish their definitions and bill by them. The Standard applies a stricter definition to the same data. A conversation can be billed correctly under your contract and still fail the Standard.
Why do repeat contacts count as failures? Won't some be false positives?
Some will. A customer may come back about a different order, a different feature or a follow-up. That is why the window is adjustable (default 7 days), why a result is shown for 1, 3, 7, 14 and 30 days, and why there is an optional same-topic filter that requires the two conversations to be textually similar. Turning the filter on reduces false positives and may miss repeat contacts whose wording differs. Treat repeat contact as a flag to sample, not a finding.
More answers on the FAQ page. The full rules and their known limits are in the Standard.