The creative strategist workflow, from research to learnings

Run creative testing as one loop, in the same order every week. Research feeds a batch, the batch gets a brief, the test gets a verdict, and the verdict shapes next week's research.

The weekly loop in one view

Most creative strategy workflows break in the same place: the end. Ads go live, results get glanced at, and nothing is written down in a form the next brief can use. So build the loop around the verdict, not around the launch.

Here is a week that works for a team shipping new Meta ads every week. Move the days around, but keep the order.

• Monday: research. Update desires, customer phrases and competitor ads. Decide what is proven and what is a guess.
• Tuesday: pick next week's batch and write the brief. One concept, one ad set.
• Wednesday: pre-flight the brief, then hand it to the creator or editor.
• Thursday and Friday: review cuts, name the ads, launch.
• Seven days after launch: give the batch its verdict and log what it says about the desire, angle, awareness stage and format.
• Always running: the tracker of past batches, so Monday's research starts from results instead of opinions.

Step 1: Research that ends in a ranked list

Research is not a folder of screenshots. It should end in a short ranked list of desires for each product, with one core desire at the top. For each desire, write who has it, what market they are in, and the mechanism that makes your product deliver it.

Then mark every line as proven, likely or assumption. Proven means an ad or an order already showed it. Likely means reviews, support tickets or interviews point to it. Assumption means it is your idea. This one habit stops a team from briefing a whole month around a hunch.

Pull customer language from reviews and count it. When the same sentence keeps turning up in different words from different buyers, tally how often. The most repeated phrase is your hook candidate, in the customer's voice. Competitor ads belong here too, but use them to spot angles that are crowded, not to copy.

Step 2: Define the batch before you brief anything

A batch is one concept tested in one ad set. Give it a number. Batch 41 is a thing you can find later. 'That video from March' is not.

Before you write a word of the brief, fix four tags for the batch: the desire it speaks to, the angle (the specific way into that desire), the awareness stage of the viewer and the format. Those four tags are what let you ask later which desires win, which angles stall and which formats carry. Skip them and every result is a one-off story.

Change one thing at a time where you can. If batch 41 uses a new angle, keep the format familiar. If you test a new format, keep the angle you already know. Otherwise a Miss teaches you nothing.

Step 3: Write the brief in three parts, then pre-flight it

Split the brief into strategy, creative and production. Strategy is the desire, the person, the angle, the awareness stage and the proof you have. Creative is the hook, the story beats and the ask. Production is what the creator needs to shoot: shot list, length, props, deliverables. Editors and creators should not have to read the strategy to find their instructions, and the director should not have to read the shot list to judge the idea.

This is also where a creative director workflow earns its keep. The director's review happens on the brief, before filming, not on the finished cut. Changing a sentence costs nothing. Reshooting costs a week.

Run a pre-flight before anything is filmed. Common failures:

• A claim with no proof behind it.
• Three hooks that all make the same promise.
• A landing page that does not continue what the ad started.
• No clear desire or awareness stage on the brief.
• A format the creator has never shot, with no reference.

Step 4: Name the ads so results find their way back

The most boring step decides whether you learn anything. Every ad name should carry the batch number, desire, angle, awareness stage, format and creator. If the name is 'Final_v3_new', the result cannot be sorted by anything, and your hit rate by angle is impossible to calculate.

Hitrate writes the ad names for you from the batch, so each result maps back to its tags without anyone retyping them on a Friday afternoon. Without a tool, keep one naming pattern in a shared doc and never deviate from it.

Launch all ads of the batch together, in one ad set, with the same budget logic you use for every test. Ads pushed from Hitrate arrive in Meta paused, so someone still makes the final call to switch them on.

Step 5: Give every batch a verdict after 7 days

Decide the rule before you see the numbers. Here is the one Hitrate uses, applied per campaign, not across the account.

A batch is a Scaler if it took a big share of its campaign's spend and the campaign spent at least 10% more per day than the week before. It is a Spender if it took a big share but the campaign did not grow. It is Efficient if it beat the campaign's own ROAS but got little spend. It is a Miss if it did neither.

A big share depends on campaign size: 30% for a campaign under $1,000 a day, 20% under $5,000 a day, 10% above that. Example: a campaign spends $2,000 a day, so a big share is $400. Batch 41 is taking $520 a day. The campaign was at $1,700 a day the week before, so it is up 17.6%. Batch 41 is a Scaler.

Judge on ROAS against the campaign's own baseline, or on a target CPA if that is how you buy. Do not compare a prospecting campaign's ROAS to a retargeting one. And give late bloomers room: a batch can still become a Scaler in week 2 or 3, so keep the label open for a while rather than killing it on day 8.

Hitrate syncs results daily from Meta, read-only, and applies this rule so no one fills in a verdict by hand. The labels and numbers are calculated, never written by an AI model. If you are not connected yet, import an Ads Manager export.

Step 6: Turn verdicts into learnings and the next brief

Hit rate is the share of tested batches that became a Scaler or a Spender. Calculate it overall, then split it by desire, angle, awareness stage and format. Example: if you tested 20 batches and 5 were Scaler or Spender, your hit rate is 25%. Now look at the splits. Perhaps most of those five came from one desire. That is next Monday's research agenda.

Efficient batches are worth a second look. They beat the baseline on ROAS but did not get spend. Often the fix is the ad set setup or the budget, not the creative, so retest before you write the idea off.

Write each learning as a single sentence tied to a batch: what you believed, what happened, what you will do next. Then upgrade or downgrade the proven, likely and assumption marks in your foundation. That closes the creative strategist workflow, and the next brief starts from what sold.

Related guides

• How to identify creative fatigue in Meta ads (with thresholds)
• Meta hook rate benchmark: what counts as a good hook rate
• How to improve hook rate on Meta video ads
• A Meta ads creative testing framework you can run weekly
• When to Kill a Meta Ad in a Creative Test

Updated 2026-10-02