Guide
One take, six ads, if you paused between the blocks
A three-minute recording contains six blocks. An opening, a problem, a mechanism, a proof, an objection answered and an ask. Say them as standalone thoughts with clean pauses and they recombine. This page covers how to record them, which variant layers to work down, and which variants are real. By the end you will get about six ads out of one sitting, with no second camera day.
- Ads a well-planned three-minute take can carry
- 6
- Extra camera days any of them require
- 0
- Export cost per output minute, per variant
- 20 CR
A sentence starting 'and that is why' cannot be moved anywhere
Record three minutes explaining a product properly. You will have said several distinct pieces. Those blocks recombine.
The proof can lead. The objection can be the hook. The explanation can become a standalone informational ad with a different opening in front of it. None of that requires more filming, only more decisions.
The mistake is treating the recording as a single linear asset producing a single linear ad. It is a library of statements captured in one sitting.
A sentence starting 'and that is why' cannot be moved anywhere. It is welded to whatever came before it. Every one of those in your take costs you a variant.
So most of the work happens before you press record. Say each block as a complete standalone thought rather than referring back to what you said a minute ago.
Pause cleanly between blocks. Those two habits separate a take that yields one ad from a take that yields six.
Ads per camera day, by how the take was planned
Same three minutes of footage. The difference is whether the blocks were said as standalone thoughts.
- One linear read1 ad
- Blocks, no clean pauses3 ads
- Blocks, planned and paused6 ads
Work down the variant layers, never across them
Order the layers by how much they move results and finish one before opening the next.
The ordering matters more than the list does. Change a little of everything at once and you get six files that differ in five ways. The result cannot be attributed to anything.
Working down the list gives you a hook test, then a length test, then a pace test. Three rules you can still use next quarter.
A hook test with a length change in it is two half-tests, and half-tests are how a year of work produces no rules at all.
Finish the hook layer before you open the length layer, even when the second one looks more interesting on a Monday morning.
- The opening. Different first sentence, same body. Biggest effect, lowest cost, always first.
- Length. A short cut keeping hook, proof and ask, against a longer cut keeping the explanation.
- Emphasis. Lead with the objection rather than the benefit by reordering the surviving lines.
- Pace. A tighter, faster edit against a more relaxed one changes who stays and who leaves.
- Coverage. Where the footage sits, and how much of the runtime it occupies.
- Caption style and music. Real variables, smaller effects, worth testing occasionally rather than every batch.
This is the whole editor
Highlight a phrase and a clip lands on those exact words. No timeline, no keyframes, no layers.
Ten files that are one idea is one shot on goal
A different colour on the captions is not a test. Neither is a different track under the same edit, nor a small trim at the end.
The test for whether a variant is real: could it plausibly perform differently, and would you learn something transferable if it did. If either answer is no, you made another file rather than another experiment.
This matters because cosmetic variants inflate your production count while doing nothing for your win-rate arithmetic.
The same file at three aspect ratios belongs in the same category. Necessary for placements. Worth nothing as a test. One row in your log, not three.
Log it as three and by March you will believe you tested thirty ideas when you tested ten. That miscount is what makes a thin quarter look like a mystery.
A real variant against a repaint
One row of your log per idea, not per file. The miscount is what turns a thin quarter into a mystery.
| A real variant | Another file | |
|---|---|---|
| What changed | The first sentence, or the order of the argument | The caption colour, or the music bed |
| Could it perform differently | Yes, and you can say why in advance | Not in any way you could predict |
| What you learn | A rule you can reuse next quarter | Nothing transferable |
| Rows in the log | One each, with a hypothesis attached | One row for the whole set |
Six from one sitting, and only the openings go live together
Record one take of up to three minutes covering the whole argument, then build from that single file in order.
The base cut first, tight, with the strongest opening you recorded. Then three more cuts identical except for the first sentence.
Then one short version of the likely winner keeping only hook, proof and ask. Then one reordered version leading with the objection. If the take contains a demonstration, one cut that opens on it with no words.
That is six ads from one sitting, each carrying a hypothesis you could write down, and none of them needed a second camera day.
Do not run them all at once. Put the opening variants live together, because that is the comparison you want. Hold the length and order variants for the following wave.
Six ads from one sitting, in build order
One take, six hypotheses. The opening variants go live together, because that is the comparison you want.
1 base cut
Tightest edit, strongest opening
3 hook variants
Same body, new first sentence
1 short cut
Hook, proof and ask only
1 reorder
Objection leads instead
Wave two
Length and order, next week
Every one of the six costs minutes, which is why the six happen
This only works if each variant costs minutes. That is what Cutroom is for. The take is transcribed, the edit is drafted, and you build each variant on the transcript.
Delete a line you do not want and the cut rebuilds around it. Change the pace and it re-cuts. Trim the opening or the ending. Then export a finished 9:16 MP4 at 20 credits per output minute.
The sixth variant costs what the first did. An editor prices each reorder as a fresh job, and that is the reason most teams stop at two.
One ceiling decides whether your last recording is usable. Three minutes of source is the cap, so a take that rambles for eight minutes gets recorded again rather than trimmed down.
Plan the blocks before you press record and that ceiling never bites. The rest of the facts are fixed. Somebody is on camera. Every export is a 9:16 MP4. There is no timeline underneath.
The six hypotheses are still yours to write. A library of statements is only a library if you said six separate statements worth keeping.
Questions people ask
- How many usable ads should one take produce?
- From a well-planned three-minute take, six to eight variants that each carry a real hypothesis is a reasonable target. Beyond that you are into cosmetic differences that inflate the count without adding information.
- Should repurposed variants run at the same time?
- Run the opening variants together, since that is the comparison you want. Hold the length and structure variants for a second wave, so the delivery system is not splitting budget across a dozen near-identical candidates.
- Does reusing the same body hurt if the audience overlaps?
- Recognition happens in the opening frames, so a new opening buys real attention even on a familiar body. It has a limit. Once several openings have been seen, the body has been consumed and it is time for a new recording.
- How long should the source take be?
- Three minutes is a good ceiling. Long enough to cover the argument with a few alternate phrasings, short enough that reviewing and cutting it does not become the bottleneck it was meant to remove.
- Who cannot use this at all?
- Anyone whose source is a silent montage, a screen recording or a multi-camera shoot. The method needs continuous speech from one person. If that is not what you have, use an editor and a timeline, or book a creator to record the blocks for you.
One sitting, one body, six openings. Each comes back as a finished vertical ad for the price of an export, and that is where cheap creative volume actually comes from.