Pay-Per-Call Buffer: What It Is and How to Set One

By the Tagloud team · November 13, 2026

Almost every pay-per-call deal has a buffer, and almost every payout dispute touches it. A pay-per-call buffer is the minimum length a call must reach before the buyer pays for it. This guide explains how the buffer and billable duration work, how buyers and publishers agree on them, how to choose a buffer for your vertical, and why a buffer works best when it is paired with written qualification rules.

It is written for publishers and affiliates, and is just as useful for buyers and networks.

This is general marketing guidance, not legal, insurance or compliance advice. If you work in regulated verticals such as SSDI, legal intake, insurance or Medicare, your buyer's written terms and your own counsel come first.

What is a pay-per-call buffer?

The buffer is a time threshold. A call that ends before it is not billable; a call that reaches it is. If a campaign has a buffer of, say, 90 seconds, a 45-second call pays nothing and a 95-second call pays the agreed payout.

It exists because many calls end almost immediately: misdials, hang-ups during the greeting, wrong companies. Both sides agree that anything under the threshold simply doesn't count.

Buffer vs billable duration

The two terms are often used interchangeably, but they differ:

• Buffer: the threshold itself, such as "90 seconds".
• Billable duration: the part of the call that is measured against the buffer.

The second one causes more trouble than the first. Does the clock start when the caller dials, when an IVR answers, or when the call connects to the buyer's agent? Do hold and queue time count? Two people can agree on a 90-second buffer and still disagree about whether a call passed it, because they are measuring different things.

How buyers and publishers agree on a buffer

The buffer is part of the commercial terms, next to the payout. A clean agreement covers four points:

1. The threshold, in seconds, for each campaign.
2. Where the clock starts: at connection to the buyer, or earlier.
3. Whose numbers count. Usually the call platform that routes the call is the shared record, so both sides look at the same duration.
4. The duplicate window: how long after a first call a repeat call from the same number is not paid again.

Write these down per campaign. A buyer may run several offers with different buffers, and a publisher needs to know which threshold applies to which number.

How to choose a buffer per vertical

There is no single right number. It depends on how long a real caller takes to get to the point. Ask three questions:

• How long is the greeting and routing? If callers pass through an IVR or a screening question before reaching an agent, a short buffer may pay for calls where no conversation happened.
• How quickly does a real conversation show intent? A caller booking a simple service may state what they need in a few seconds. A caller describing an injury, a disability claim or an insurance situation usually needs longer.
• What is the payout? Higher payouts usually come with longer buffers, because the buyer wants more certainty before paying.

As an illustration, not a benchmark: a buyer might set a shorter buffer for a quick-booking home service offer and a longer one for a legal intake offer where the first minute is spent on greeting and basic questions. The right values come from listening to a sample of your own calls and noting when real conversations usually begin.

What a buffer can't tell you

A buffer measures time, not quality. A call can pass any buffer and still be worthless to the buyer:

• An existing customer calling about a bill
• A caller outside the buyer's service area
• Someone who already has an attorney or a provider
• A long hold before anyone spoke

It can also cut the other way. A qualified caller who is transferred quickly, or who calls back after a dropped line, can fall under the buffer. We cover this in more detail in call duration is not call quality.

For publishers, this matters: a buyer relying only on the buffer may still reject calls later, with reasons that were never agreed. That is where disputes start.

Pairing the buffer with qualification rules

The buffer works best as the first filter, with written qualification rules as the second. The buffer removes the obvious noise for free; the rules decide which of the remaining calls are worth paying for.

Good rules are short, specific and checkable on the call: "lives in one of these states", "is not already represented", "wants a quote". Each rule should be one fact the caller can confirm out loud. Agreeing on rules up front means a rejected call comes with a reason both sides accepted in advance, and a quote from the call to back it up. That keeps disputes over junk calls short.

Some buyers pay on transfers instead: if the call reached their agent, it counts. That works too, as long as everyone knows it applies.

Step by step: setting a buffer

Step 1: Listen to a sample of calls

Review a sample of calls for the campaign. Note how long greeting and routing take, and when callers first show real intent.

Step 2: Agree on billable duration

Settle where the clock starts and which system's duration both sides will use.

Step 3: Set the threshold per campaign

Make it long enough to skip the noise you heard in the sample, but not so long that quick, genuine calls are lost. Set the duplicate window too.

Step 4: Add written qualification rules

List the facts that make a call worth paying for, or agree that transfers count as qualified. Share the list with every publisher before traffic starts.

Step 5: Review after real traffic

After a few weeks, look at calls just under and just over the buffer. If many good calls are falling short, the buffer may be too long. If many paid calls fail the rules, the rules are what protect the payout. Compare sources on cost per qualified call, not cost per call.

How Tagloud helps

Tagloud pulls every call from CallGrid, Ringba, Retreaver or any webhook, with call length, payout, publisher and buyer. Each call gets a verdict, either by AI against your own qualification rules or as "transferred = qualified", with a reason and quotes from the call. Short calls, hang-ups, repeat callers, wrong numbers and out-of-area callers are flagged, and every publisher is ranked by cost per qualified call. On the publisher side, that means you see the same verdict the buyer sees and know which traffic to scale.

Want to see how a call holds up against a set of rules? Try the free call checker with a transcript or recording.

FAQ

Is a longer buffer always better for buyers?

No. A longer buffer filters more noise but can drop genuine callers who were helped quickly. Rules catch bad long calls better than extra seconds do.

Should publishers ask for a shorter buffer?

A shorter buffer only helps if the calls also meet the buyer's rules; otherwise they may be rejected later. Clear rules are usually worth more than a few seconds off the buffer.

What if the buyer and publisher see different durations?

Agree in advance which system is the shared record, usually the platform routing the call, and where the clock starts. Most mismatches come from measuring different parts of the call.

Does a call that passes the buffer have to be paid?

That depends on your agreement. If the terms say any call over the buffer is billable, it is. If they include qualification rules, a call must pass both. Put the answer in writing before traffic starts.