A workable video ad production workflow for one person fits into about six hours a week, split across five days: brief on Monday, generate on Tuesday, batch review on Wednesday, launch on Thursday, read the results the following Monday. Three steps carry the whole thing, and they are the brief, the batch review and the weekly read. Storyboards, per-ad approval rounds, custom thumbnails and a separate QA pass are all optional, and the week runs better without them.
What a week actually looks like, hour by hour
This is the shape I would defend for someone running one active account and a weekly test cadence. Times are the block you protect on the calendar, not the time the work takes if it goes badly.
| Day | Block | Time | Output |
|---|---|---|---|
| Monday | Read last week's results, pick the variable to test | 45 min | One sentence: what changes this week |
| Monday | Write the brief | 30 min | Offer, audience, angle, constraint, format |
| Tuesday | Generate the batch | 45 min | 10 to 14 outputs from one brief |
| Wednesday | Batch review, all outputs in one sitting | 60 to 70 min | 6 to 9 approved, the rest cut or regenerated |
| Thursday | Upload, name, structure the ad set, launch | 45 min | Live ads before 14:00 |
| Friday | Nothing | 0 | Deliberately nothing |
That is roughly four and a half hours of protected time, and it usually runs to six once something goes wrong. Friday is empty on purpose. A creative launched on Friday collects weekend data with nobody watching it, and the Monday read starts with a cleanup instead of a decision.
Why the cadence is weekly rather than monthly
Weekly is not an arbitrary rhythm. Our 2026 ad fatigue benchmark puts CTR decline at 15 to 20 percent across a creative's first two weeks, with week three arriving as a cliff at 45 to 70 percent below the launch baseline. By week five a creative averages 38 percent below its peak. If your refresh cycle is monthly, the second half of every month runs on dead creative.
The same benchmark reports that brands shipping 15 to 50 variants a month see three to five times longer campaign lifespan than quarterly refreshers. One weekly batch of six to nine published ads lands inside that band. Decay speed does vary by vertical: food and beverage hits a 40 percent CTR decline in about 9 days, fashion in 12 to 14, beauty and DTC around 18, electronics around 21, B2B SaaS around 28. A SaaS team can genuinely run fortnightly. A food brand cannot.
Which steps you can drop
Most published workflows are additive, and a team of one cannot afford additive.
Drop the storyboard. For a 15 to 30 second ad built from footage you already own, a storyboard describes a video that already exists in the library. The brief plus a shot list does the same job in a tenth of the time.
Drop per-ad approval. The largest single saving in the week, covered below.
Drop the second opinion on hooks. One person reading eight hooks against a written constraint list is faster and no less accurate than two people discussing them. Hook selection is a job for the test, not the meeting.
Never drop the brief. An unwritten brief produces outputs that vary on everything at once, which makes the weekly read meaningless. The brief is what turns ten files into a test.
Never drop the weekly read. Producing without reading is just spending. Our creative testing workflow walks the read step in more detail than a calendar block can.
Why batch review beats per-ad review
Per-ad review is where solo workflows die. Reviewing each ad as it is produced means reloading the brief, the brand constraints and the platform rules once per ad, and every reload costs minutes.
Batch review inverts it. Open every output from one brief in one sitting, hold the constraint list once, and make three decisions per ad: publish, regenerate, or cut. Budget about five minutes of review per output. Twelve outputs is an hour, which fits in a single Wednesday block.
Two rules keep the batch honest. First, review against the brief and the constraints, not against your taste. Second, cap the regeneration loop at one pass. An ad that needs a second regeneration is an ad the brief failed to specify, and the fix belongs in next Monday's brief.
Batch review also gives you the number that governs your real cost per published ad: keep rate, meaning published divided by produced. If nine of twelve outputs go live, your keep rate is 75 percent, and shipping eight ads means producing eleven. Everything you spend scales off the produced number, not the published one, which is why keep rate is the production metric worth tracking before you optimise anything else in the week.
The numbers to report weekly
Five lines. If the weekly report takes longer than ten minutes to fill in, it will not get filled in.
| Metric | How to get it | What it tells you | Where it should sit |
|---|---|---|---|
| Produced | Count of outputs generated | Your true cost base | Whatever the cadence needs |
| Published | Count that went live | Actual throughput | 6 to 9 for one account |
| Keep rate | Published / produced | Brief quality, not editing quality | Above 60 percent |
| Hook rate on new ads | 3-second views / impressions | Whether the top of the ad works | 20 to 30 percent on TikTok in-feed, 15 to 25 percent on Meta feed |
| Frequency on the winner | Platform report | How long the winner has left | Above about 3 means fatigue has arrived |
Notice what is absent: CPA on new creative in its first four days, and any ranking of ads by ROAS inside a week. Both are noise at that sample size, and both will talk you into an early kill.
A rising keep rate with flat published volume means your briefs improved. Flat keep rate with rising produced volume means you bought throughput without fixing the brief, and you paid for it twice.
What this costs, and where the money goes
The credible comparison is per published ad, not per file. In-house editing typically runs 30 to 90 minutes an ad, which at a loaded hourly cost of about EUR 45 (salary plus overhead, not just salary) puts a single ad somewhere around EUR 25 to 70 of your time. Agencies commonly quote EUR 100 to 400 per ad. Genyad charges 1 credit per variation, about EUR 1.52 on the EUR 99 Growth pack, plus the roughly five minutes of review each output still needs. We make Genyad, so weigh that accordingly, and the honest framing is that generation cost drops toward zero while review time does not. Our video ad cost calculator will take your own hourly figure and keep rate.
Run the arithmetic on a 60 percent keep rate and 40 published ads a month: you produce 67. At EUR 45 loaded and an hour an ad, that is roughly EUR 3,000 of internal time. Cheap generation moves most of that to review, and review is around 5.6 hours at five minutes each. It does not move all of it, and any tool that claims otherwise is selling you the produced number rather than the published one.
Where this workflow breaks
Two failure modes, both of which I have caused.
The library runs dry. A footage-first workflow needs new source material, and a library that has not been added to in three months starts producing variations that look like each other. Book one shoot day or one bulk screen-recording session per quarter.
Two accounts, one person. The calendar above holds for one active account. At two, Wednesday's review block doubles and Monday's read splits, and something has to give. Agencies solve this with a different shape entirely, which is what our agency creative workflow covers.
Frequently asked questions
How long does one video ad take to produce?
Manual editing of a single short-form ad from existing footage typically takes 30 to 90 minutes, depending on how much of the source material you have to hunt for. With generation from a tagged library, the production step drops to minutes and the remaining cost is review, which runs about five minutes per output. Budget by the batch rather than by the ad.
How many video ads should one person publish per week?
Six to nine published ads a week per active account is a realistic target for a solo operator, which puts you at 24 to 36 a month. That sits inside the 15 to 50 variants a month band our fatigue benchmark associates with materially longer campaign lifespan. Publishing more than that alone usually means review has been skipped rather than sped up.
Can this workflow run without a creative brief?
It can produce files, but it cannot produce a test. Without a written brief every output varies on hook, offer, angle and format at once, so the weekly read cannot attribute a result to anything you did. A five-field brief takes 30 minutes and it is the highest-leverage half hour in the week.
What should I cut first if the week runs short?
Cut Thursday's polish, not Wednesday's review. A badly reviewed ad costs you spend, while a missing ad costs you one week. If the whole week collapses, let last week's winner run on rather than shipping unreviewed creative.