The short version
- Run ad creative as a weekly loop with fixed days: learn, brief, make, review, launch paused, then start again.
- Give every stage one owner, even in a team of one. Where you can, the person who approves an ad shouldn't be the person who made it.
- Make ads in batches by concept and review them in one sitting against a written checklist, so nothing waits in a chat thread.
- Leave time for Meta's review, which Meta says is typically done within 24 hours but can take longer. Launch on Thursday, not late on Friday.
- Keep a three-line log of every test. Next week's brief starts from it, not from a blank page.
The short answer
A small team gets the most ads from the fewest hours with a weekly loop of five stages, each with one owner and a fixed day: learn, brief, make, review and launch. In the worked example at the end, two people brief, make, check and launch 20 ads in about nine hours of combined time.
The loop matters more than the tools. Small teams rarely run out of ideas. They lose days to waiting: for a brief nobody wrote, a review nobody owned, a launch that sat in drafts. Fixed stages and days take the waiting out.
Who does what in a team of one to three
There are four jobs in ad creative, whatever the size of the team. In a large team they're four people. In a small one they're four hats.
| Job | What it covers | What it hands on |
|---|---|---|
| Strategist | Reads last week's results, picks the angles, writes the brief | A one-page brief |
| Maker | Makes the ads, fixes details, writes or checks the copy | A batch of draft ads |
| Reviewer | Checks every ad against the checklist, kills or approves | An approved set |
| Launcher | Puts the ads live, names and tags them, watches the first days | Live tests you can read |
How to split them:
- One person: wear all four hats, but on different days. Brief on Monday and review on Wednesday, so you see the ads with fresh eyes.
- Two people: one is strategist and reviewer, the other maker and launcher, so the approver never made the ad.
- Three people: split the maker from the launcher, so one person owns everything inside the ad account.
In BulkAds the same split can be set as roles. Teammates join an account as an admin, an editor or a viewer. Editors can make ads but can't launch them or touch Autopilot, admins can do everything except billing, and viewers can only look. Team seats start on the Agency plan, which includes unlimited seats. The Creator and Pro plans are for one person.
The weekly loop, day by day
| Day | Stage | Owner | Time box |
|---|---|---|---|
| Monday morning | Learn: read last week's tests, update the log | Strategist | 45 minutes |
| Monday afternoon | Brief: the question, the concepts, the claims you may use | Strategist | 45 minutes |
| Tuesday | Make: batches by concept, first pass on copy | Maker | 3 to 4 hours |
| Wednesday | Review: one sitting, the checklist, then fixes | Reviewer, then maker | 1.5 hours |
| Thursday | Launch: build paused, check, approve | Launcher | 1 hour |
| Friday to Sunday | Run: hands off, apart from stopping clear losers | Launcher | 15 minutes a day |
The time boxes are sized for about 20 ads, the number in the worked example. The order is the real point: each stage takes its input from the one before, and nothing starts without it. If the brief isn't written by Monday evening, Tuesday isn't a making day.
Launching on Thursday leaves room for Meta's review. As of October 2026, Meta's Advertising Standards (opens in a new tab) say review starts automatically before an ad runs and is typically completed within 24 hours, though it can take longer. Its Marketing API reference for ads (opens in a new tab) adds that new ads stay pending and don't run until Meta approves or rejects them. A Thursday launch gives a slow review Friday to finish, and the ads three or four days of delivery before Monday's reading.
The loop is for steady weeks. A product launch or a sale date you can't move needs a bigger, one-off round, which our guide to the creative sprint plans day by day.
Stage 1: a brief that fits on one page
The brief is the cheapest place to fix an ad. Ten minutes deciding what to test saves hours of making ads nobody needed. A useful one fits on a page:
- This week's question, in one sentence, such as "does a price-per-meal angle beat a taste angle for the starter pack?"
- One product, one offer, one landing page. Mixing them makes the results unreadable.
- Three to five concepts, each a different angle, format or setting. Our guide to concept vs iteration testing covers how to balance new ideas against versions of a winner.
- Approved claims. The exact proof points the ads may use, and nothing else.
- Count and sizes, for example four concepts with five executions each, all at 4:5, and 9:16 versions of the best later.
- What to avoid. Angles that lost last month, words a lawyer flagged, any competitor's look.
The creative brief template has a fill-in version, the free creative brief generator drafts one for your product, and the guide to writing an ad creative brief for AI covers the extra detail AI tools need.
Stage 2: make in batches, by concept
Make one concept at a time, not one ad at a time. A batch per concept keeps each angle consistent inside its batch and the differences clear between batches, which is exactly what a test needs to be readable.
- Start every batch from the brief, not from memory. Paste the concept and the approved claims straight into the tool you use.
- Make a few more than the brief asks for. Some will fail review, and it's quicker to choose from a spare than to make one more on Thursday.
- Fix small, don't restart. A wrong word in a headline is a two-minute fix. Regenerating the whole ad for it wastes time and can lose the version you liked.
- Write the copy with the image, not after it. Headline and primary text should come from the same concept as the picture.
- Save as you go into the week's folder, one subfolder per concept.
In BulkAds a batch makes up to 12 ads from your product photo, each with its own headline, primary text and description, and your own instructions can steer it. Kettlewick's 20 ads could be four batches of five, one per concept, using 40 credits in all. If a headline is slightly off, regenerate just that element, not the whole ad.
One honest limit: BulkAds' video ads add motion to a product scene. A concept that needs a person talking to camera still needs a creator, and their video runs from Meta Ads Manager, while BulkAds makes the stills and AI video variations around it.
Stage 3: review in one sitting
Review leaks time when it happens in pieces: a message here, a comment there, a "looks fine?" that waits a day. Do it in one sitting, against a written list, and decide every ad there and then.
| Check | Kill or fix if |
|---|---|
| Product | Shape, label or colour differs from the real thing |
| Words on the image | Any letter is wrong, or a price or number is unclear |
| Claims | Anything appears that isn't on the brief's approved list |
| Personal attributes | The ad states or implies something about the viewer, such as their health or money worries |
| Safe zone | Key text sits where the 9:16 screen furniture will cover it |
| Copy match | The headline and primary text say something different from the image |
| File name | The name doesn't show concept and version |
The personal attributes line comes from Meta's ad standards on personal attributes (opens in a new tab), checked in October 2026, which don't allow ads that assert or imply things like a person's health, vulnerable financial status or age. On Meta's page, "Do you have diabetes?" is a banned example, while "New diabetes treatment available" is allowed. For Kettlewick, "Watching your salt intake?" points at the viewer, while "Spice blends with no added salt" talks about the product. The safe zone checker shows where 9:16 text gets covered, and the ad design checklist goes deeper on layout.
Three rules keep the sitting short:
- Anyone can kill an ad, one person approves. Debate is for the brief, not the review.
- Fix or bin inside the sitting. "Let's come back to it" is how Thursday's launch slips to next week.
- Note why ads die. If half a batch fails for the same reason, fix the brief or the product photo, not the ads.
BulkAds runs a pre-flight check on every ad before you see it, scoring product match, on-image spelling, brand fit and visual quality and flagging likely Meta policy risks. It shortens the sitting. It doesn't replace your own read, least of all on claims.
Stage 4: launch paused, then approve
Launch in two steps: build everything paused, look once more inside the ad account, then switch it on. The second look catches errors that only show in context, such as a caption on the wrong image, a link to the wrong product or a 9:16 ad sent to a feed-only placement.
Keep each week's ads together so they compete on the same terms and can be compared. Our guide to how many ad creatives to test covers how many to put in one test. Name every ad by concept and version and tag its link, so your store's analytics can tell the ads apart. The naming conventions guide and the free UTM builder handle the details.
BulkAds' bulk ad launcher builds the campaign, the ad set and one ad per creative in your own Meta ad account, all paused, with UTM tags and names by angle and ID. Nothing spends until someone with launch rights approves it. Autopilot's hands-off mode only covers the tests Autopilot launches itself, never the ads you launch here.
Stage 5: read the results and write it down
The loop closes on Monday morning with a short read of last week's test and three lines in a shared log:
- What we tested: the question from the brief.
- What happened: which concepts reached a verdict, and which way.
- What we'll do next: more of a winner, a new angle, or a retest of an idea that never got a fair run.
Agree the verdict rules before the test, so Monday's reading isn't a debate. When something wins, the next brief is about iterating on a winning ad: new hooks, openings and formats of the same idea.
In BulkAds, creative analytics groups results by angle, hook style, format and size for ads made and launched there, and the results of campaigns you launch sync back to BulkAds, so much of the log writes itself. If you hand tests to Autopilot instead, its verdicts land in a test ledger, and in its default ask-first mode it proposes a budget step for a winner rather than taking it, and drafts four fresh variants of each winner.
Files and names that survive a busy week
A small team can't afford a hunt for the right file. Four habits prevent it:
- One folder per week, with a subfolder per concept and the brief saved at the top.
- An asset library of approved packshots, logos and lifestyle photos, so nobody makes ads from last year's label.
- A swipe file of ads you admire, tagged by angle, for the strategist to raid on Monday.
- One name pattern everywhere: week, product, concept, version and size, such as W44_starter_timer_v3_45, used for files, ads and the log alike.
Where small teams lose time
- Briefing in chat. The brief ends up spread across twenty messages and nobody can find the approved claims. Keep it in one document.
- Making one ad at a time. Each ad gets its own little brief, review and upload. Batches remove that overhead.
- Open-ended review. Without a checklist and a sitting, review drifts across days. Book the hour.
- Launching late on a Friday. Review and the first delivery happen while nobody is watching. Thursday gives you a working day to fix a rejected ad.
- Testing near-duplicates. Five colour swaps of one ad are one concept. Meta's engineers wrote on 2 December 2024 (opens in a new tab) that more than a million advertisers had used its generative AI tools to make more than 15 million ads in a month. Volume is normal now. Variety is what a small team can still win on, as our guide to Andromeda and creative diversity explains.
- No log. Without one, teams quietly retest ideas that already lost.
Worked example: a two-person week
Kettlewick's team is the founder, who acts as strategist and reviewer, and one marketer, who makes and launches. Here is their week for the 20 ads in the brief above:
| Stage | Who | Hours |
|---|---|---|
| Learn and brief | Founder | 1.5 |
| Make four batches of five and fix details | Marketer | 4 |
| Review in one sitting | Founder, then marketer for fixes | 1.5 |
| Launch paused, check and approve | Marketer, founder approves | 1 |
| Daily checks from Friday to Monday | Marketer | 1 |
| Total | 9 |
Nine hours for 20 launched ads leaves most of the week for everything else a small business does. For comparison, briefing a freelance designer for 20 static ads at, say, $15 an ad would cost $309 and several days of back and forth.
That isn't an argument against designers. A good designer or photographer brings craft a batch tool doesn't, and is the right call for a brand shoot, a launch hero image or packaging photos. A split that works: people for the assets you'll use all year, fast batches for weekly ads that only need to answer one question. See how many ads to make each week for setting your own volume.
Questions people ask
What is an ad creative workflow?
It's the repeatable set of steps a team follows to turn an idea into live, tested ads: learning from the last test, briefing, making, reviewing and launching. Written down with owners and days, it stops ads waiting between steps.
How many ads can a small team make each week?
In the worked example in this guide, two people brief, make, review and launch 20 new ads in about nine hours of combined time. Our guide to how many ads to make each week suggests starting at 10 to 20 for an in-house team of two or three, then adjusting to your own making limit and to what your spend can test.
Who should approve ads in a small team?
One named person, ideally not the one who made the ads. Anyone can flag or kill an ad in review, but a single approver keeps the sitting short and makes it clear who owns what goes live.
How long does Meta take to review a new ad?
As of October 2026, Meta's Advertising Standards say review is typically completed within 24 hours but can take longer, and new ads don't run until Meta approves or rejects them. Launching on a Thursday leaves a working day to fix a rejected ad before the weekend.
Can one person run this workflow alone?
Yes. Wear all four hats but on different days, so you brief on Monday and review on Wednesday with fresh eyes. A written checklist matters even more when nobody else is checking your work.
Can my team share one BulkAds account?
From the Agency plan up, yes. Teammates join as admins, editors or viewers. Editors can make ads but not launch them, so the launch stays with the owner and the admins you choose. The Creator and Pro plans are for one person.
Keep going
- Features
AI ad generator
On-brand image ads in batches of up to 12
- Features
Bulk ad launcher
A whole batch live on Meta in one go
- Features
Creative analytics
Which ad actually sold — and why
- Free tools
Creative brief generator
A one-page ad brief from short answers, with checks, ready to copy
- Guides
Creative brief template for ads, with three filled-in examples
A ten-field creative brief template for image and video ads, three filled-in examples for a static, a creator video and an AI batch, policy notes and a send checklist.
- Guides
Creative sprint: plan, make and launch a round of ads in five days
A creative sprint turns one week into a round of launch-ready ads. The day-by-day plan, who does what, a worked example for a sock brand and the review gate before launch.
Written by the BulkAds team. Published , last checked . Meta changes its tools often — if something here no longer matches Ads Manager, tell us at contact@bulkadsai.com.
BulkAds AI is an independent product and is not affiliated with, endorsed or sponsored by Meta Platforms, Inc. Facebook and Instagram are trademarks of Meta Platforms, Inc.