Duration is the wrong thing to sell

Almost every recorded tier in this trade is sold by the minute, and almost none of the cost sits in the minutes. The expensive part of a recorded assessment is everything that happens before and after the recording, and none of it scales with duration in the way the price list implies.

Sell a five minute video and a ten minute video at double the price, and you have told the buyer that the second one contains twice as much of whatever they wanted. You have also told yourself to keep talking.

Where the time actually goes

Time one recorded job properly, from opening the brief to sending the file, and the breakdown is consistently lopsided.

The only row on that table that grows with the length of the video is the row that costs you the least. Everything else is a per-job overhead you pay identically for four minutes and for twelve.

This is the same arithmetic that makes flat platform fees so punishing at the bottom of a rate card, worked through in the piece on where a flat fee sets your floor. Fixed costs do not care how big the job is, so they decide the price of small jobs entirely.

What per-minute pricing does to the work

A per-minute price is a piece rate, and piece rates work: when Safelite Glass moved to piece-rate pay, Edward Lazear's 2000 study in the American Economic Review measured "a 44-percent increase in output per worker". A piece rate reliably produces more of whatever is counted, and here what is counted is minutes, so it gets gamed even by honest people who are not trying to game anything.

You start narrating what you are about to say before you say it. You restate the brief back at the buyer for thirty seconds. You avoid the tight cut that would have made the middle section better, because the tight cut costs you money.

None of that is fraud. It is what happens when you attach the fee to the thing that gets longer when you work less well. Editing, the one activity that reliably improves a recorded deliverable, is under a per-minute rate an activity that reduces your own invoice.

Buyers notice the result before they can name it. A recorded review that arrives at exactly the length that was purchased, with an obvious two minutes of nothing in it, reads as bought rather than made. That perception is what shows up later in the reviews that persuade the next buyer, or fails to.

Price the scope, state the duration as a consequence

The fix is not complicated. Define the recorded tier by what it covers, and mention duration as an expectation rather than a unit.

A workable listing says what the recording addresses - the axes, the number of points, whether there is a summary at the end - and then adds "typically eight to twelve minutes". That sentence gives the buyer everything they were trying to get from a minute count, and it commits you to none of it.

It also does something useful to your own incentives. If the job is well covered in seven minutes, you send seven minutes and keep the margin. If it genuinely needs fifteen, you send fifteen and the buyer receives more than they expected at no extra charge, which is a cheap way to buy goodwill in a trade where goodwill converts into repeat bookings.

The equivalent argument for written work, where the temptation is word count rather than runtime, is in the case for structured scoring over prose. The pattern is the same: attach the price to the thing the buyer wants, not the thing that is easy to count.

Guidance written for buyers tends to agree, incidentally. Rate Penis on how a commission is normally framed describes people asking for coverage and specificity, not for runtime, which is a reasonable indication that the minute count is a seller's convention rather than a buyer's requirement.

When duration does belong in the price

There is one honest exception, and it is worth naming so the argument does not overreach.

Real-time work is priced by duration because the inventory is time. A booked half hour is a half hour you cannot sell to anyone else, and that is a different economic object from a recording, which is covered separately in the piece on selling a fixed slot.

Very long recorded formats also have a real duration cost, because upload, export and storage stop being trivial and because sustained speaking is genuinely tiring. Above roughly twenty minutes, a duration component starts earning its place. Below that, it is measuring the cheapest input you have.

Tiering without going back to minutes

The reason people reach for minutes is that a rate card needs rungs, and duration is the only obviously ordered variable available.

There are better ones. Number of subjects covered. Whether a written summary accompanies the recording. Whether the buyer gets to send a follow-up question. Turnaround.

Each of those costs you something real, each is legible to a buyer in one line, and none of them rewards padding. The lowest-effort version of this is a recording plus a short written note, which is the pairing worked through in the post on bundling written and recorded work.

Equipment is the other thing people expect to matter and it mostly does not - a quiet room beats an expensive microphone, and the general treatment of what recording gear is actually worth buying sits with the measurement site's equipment notes rather than here. The automated tools most buyers have already tried before they commission a person produce output instantly and at no marginal cost per minute, which is documented on the tools side and is precisely why competing on volume of output is the wrong axis.

If you want to see how a platform actually presents a recorded tier to buyers, including where duration appears and where it does not, Rate Cock's judge listings are the closest available reference point. Read a dozen of them and count how many sell coverage against how many sell time.