It does not take your craft work. It takes variant four.

Sort your last two months of invoices by hourly yield.

The bottom of that list is not the campaign you are proud of.

It is social cutdowns, caption passes and hook variants for a client testing openings.

That work is repetitive by design. Same body, different first eight seconds, four times.

Captions get rebuilt each round, because pulling a line shifts everything after it.

It pays badly because the client sees it as one video with small changes.

You see it as four timelines and cannot say so without sounding difficult.

This is aimed at exactly that job and nothing above it.

The clients commissioning it are not going to stop needing it. Their creative testing is expanding.

So the choice is whether that volume arrives as timeline hours, or as something you deliver in an afternoon and price as a test set.

The reason a variant is cheap here is structural.

The edit is described in word indices against the transcript, and the compiler owns all the time maths.

So deleting a line is not a ripple delete that you then fix.

The plan changes and the cut recompiles.

Captions, b-roll placement and timings come back consistent, because nothing was hand-placed to begin with.

Highlight a phrase and a clip lands over exactly those words.

Change the pace and coverage shifts within a cap of 40, 45 or 52 percent depending on how hot you set it.

For a working editor the useful comparison is the shape of a revision.

In a timeline, pulling one line is a local change with global consequences that you then chase down the track.

Here the consequences are recomputed rather than chased. That difference is the entire reason variant five costs minutes.

It also removes the class of error that eats a revision round. No caption left one frame behind a cut.

The b-roll is searched and placed for you against the words, from millions of free clips, and your own uploads sit in the same picker.

A finished thirty second cut is about 110 credits. Basic is $39.99 for 2,500 credits, roughly 22 cuts.

Premium is $79.99 for 5,000, roughly 45.

That changes what you can offer.

Not one video for a day rate, but a set of openings against one body, delivered together.

The client's real problem is sample size, and nobody on that account is solving it.

Roughly 5 to 8 percent of creatives carry spend, per Motion's analysis of 550,000+ Meta ads.

An editor who can say that and deliver twelve variants is selling a testing service, not an editing hour.

Your craft work keeps its rate, because it is now clearly a different product sold to a different need.

There is a reporting angle too. If you deliver the openings, you can ask for the results.

An editor who can say which opening held is having a conversation no other supplier on that account is having.

The deliverable is a finished 9:16 MP4 with captions burned in.

There is no XML, no EDL and no round trip, so treat it as a delivery format rather than a stage in a pipeline.

Say that to clients up front. Some agencies archive project files as policy.

There is no colour work and no mixing beyond turning music on, picking the track and setting how loud it sits.

Bad room tone stays bad room tone, so brief the client on the take rather than on the edit.

The source is one talking-head take, three minutes at most, and one project yields one finished cut.

Client-supplied product footage goes into the workspace library and drops onto the phrase it belongs on.

Multi-camera and assembled sequences stay where they belong, in your timeline.

Everything vertical, spoken and repetitive belongs here instead. That is the half of your list that pays worst per hour.

Take the next brief that asks for three openings against one body. Batch the take at 100 credits.

Delete the opening in the transcript and the cut rebuilds with the same pace and the same caption pack.

Do that twice more and export each at 20 credits per output minute.

Three finished 9:16 files, captions burned in, b-roll placed, before the afternoon starts.

The variables you did not intend to change did not change, which is what makes the client's test readable.

That is the line to put in the quote. Not cheaper editing, but a result the client can act on.