Abandonment
The call ends before a defensible decision. Some of these are valid customers lost to a queue or a rigid system. You want this low.
Definition
What is a false pass in third party verification?
A false pass is a third party verification recorded as confirmed when the recording shows the customer did not agree. It is the verification version of a false positive. It counts as a success in every number a vendor reports, which makes it the hardest TPV failure to see, and it is the only one of the three TPV outcomes that carries legal exposure.
The TPV Buyer's Checklist
What to measure, and what to make any TPV vendor prove, including us.
Context
One completion rate, three outcomes
Most TPV reporting leads with a single completion or pass rate. Three outcomes hide inside it, and only two of them ever show up as problems.
Rejection
The verification reached a substantive problem: price, comprehension or authorization. That may be the verifier doing its job. A rejection rate near zero means the control never fires.
False pass
The report says confirmed. The recording says the customer did not agree. It counts as a success in every number the vendor shows you.
Why you cannot see it
Why a false pass is invisible by construction
Every TPV report starts from what the system recorded. A false pass was recorded as a pass, so it lands in the success column of the dashboard, the completion rate and the invoice. Nothing in the data marks it.
- The only evidence is in the recording, and approved calls are the ones nobody listens to.
- Pricing per completed verification pays the vendor for it.
- Pass rates are the headline number, and a false pass makes the headline look better.
Abandonments and rejections show up in reports because the system knows they did not succeed. A false pass is a failure the system records as a success. You have to go and find it.
Causes
How false passes happen
Every verification model can produce them. The mechanisms differ.
- Word matching. A verifier listening for yes hears it in “Yes, that is not correct” and passes a customer who just said no.
- Accepting a non-answer. “I guess,” “uh-huh” or a long silence treated as consent.
- Leading. A verifier, or a sales rep still on the line, steers the customer to the expected answer.
- The wrong person. A spouse or roommate who is not the account holder confirms the sale.
- Disclosure drift. A disclosure paraphrased over time until the customer is agreeing to something other than the contract.
- Improvisation. An AI verifier that answers off-script, or keeps talking until a hesitant customer says yes.
- Failing open. A system under strain that starts approving instead of stopping.
TPV audit
Two ways to find false passes
You cannot get them from vendor reporting. There are two ways in, and both start with your own calls.
Sample
Pull a random sample of your own approved calls every month. A person listens to each one and answers a single question: is the answer on this recording actually consent?
Size the sample with your compliance team, so that a false pass rate you would care about would show up in it.
Work backward
Start from cancellations, disputes and complaints where the customer says they never agreed. Every one of them started life as an approval.
Pull the verification recording behind each one and listen. A pattern by rep, by script step or by time of day tells you where to look next.
Your approvals, your audit
You will never get to audit another client's calls, and you should not need to. The approvals on your own book are your records and your risk. Ask how you would go about auditing those.
In practice that means the recording and the answer to each question have to be yours to pull, for any approval, without asking anyone.
Evaluating vendors
The questions every TPV vendor should answer
- Do you audit your own approvals?
- At what sample size?
- Who does the listening?
- How would we audit the calls you approve on our behalf?
Ask us the same questions. Every approval we return comes with its recording and the answer to each question, so you can sample our approvals the way you would sample anyone's.
FAQ
Questions about false passes
What is a false pass in TPV?
A third party verification recorded as confirmed when the recording shows the customer did not agree. It counts as a success in every number a vendor shows you, it is the only TPV outcome that carries legal exposure, and vendor reporting cannot show it to you.
What is a false positive in verification?
The same thing under another name: the verification says yes when the truth is no. In TPV it is called a false pass, an approval the recording does not support.
How do you audit TPV calls for false passes?
Two ways. Sample: pull a random sample of your own approved calls every month, and have a person listen to each one and answer a single question, is the answer on this recording actually consent? Work backward: start from cancellations, disputes and complaints where the customer says they never agreed, and listen to the verification behind each one.
Can we audit our TPV vendor's calls?
You will never get to audit another client's calls, and you should not need to. The approvals on your own book are your records and your risk. Ask how you would go about auditing those.
Can AI verifiers produce false passes?
Yes, in two ways. A system that matches words can hear the yes in "Yes, that is not correct" and pass a customer who just said no. A conversational agent that improvises can keep talking until a hesitant customer says yes. Test for the first with the contradiction test, and ask how the second is prevented and checked.
Why don't vendor reports show false passes?
Because a false pass is recorded as a pass. Reports count what the system recorded, so it lands in the success column of every dashboard, completion rate and invoice. The only evidence is in the recording.
Free download
The TPV Buyer's Checklist
Bring it to a vendor call: the number that matters, the nine numbers to make any vendor separate, the five things to make them prove, and the two questions that sort vendors fastest.
Talk to Ryan
Talk to Ryan about auditing your approvals
Tell me what you verify and roughly how many a month, and I will reply by email. If you would rather talk it through, pick a time on my calendar.
Pick a time on my calendar
Ryan Schuetz
Founder, Automatdo