I had Ads Manager open next to Shopify on a screen share with a founder. July to September they'd spent 6.6x what they did last year and only got 1.5x the orders. Only two ads ever sold at a cost that works, both videos from 2025.

Biggest issue was the landing page.

Ads went to an $80 subscription page and add to cart dropped from 12 to 18% down to 6 to 7%. I'd send everything to the bundle-builder page first, clean up the pricing, and build more variations off those two videos.

Most founders who message me on Upwork think their campaign setup is broken or Meta broke their pixel. You can tweak bids for months, but the auction only rewards ads that stop the right buyer and make a promise they believe. As a creative strategist for hire, I write five briefs a week with three variants each, and I grade them on 7-day spend share in a CBO. Here's how I run Meta ads creative strategy on live DTC accounts.

What Meta Ads Creative Strategy Covers in a DTC Account

Meta ads creative strategy covers everything before an ad hits the auction and everything I learn after buyers react to it. Media buying handles the campaign setup, pacing and account structure. I own the customer research, the angle, the hook and bridge, the visual direction and the brief your editor cuts from. I don't edit videos myself.

With DTC creative strategy, the scoreboard sits outside Ads Manager. I read Triple Whale on 7-day click, or Shopify new-customer orders if you don't have Triple Whale. I don't use 1-day view because it gives retargeting too much credit.

Platform ROAS can say 3.0x while refunds, tariffs, product cost and subscription renewals hide a loss on cold traffic. So every call I make ties back to contribution margin after ad spend and what a new customer costs you.

What an account needs also changes with daily spend.

From $0 to $100 per day, founder hustle and proving the offer matter more than ad testing. I pass on brands doing under $60K a month in revenue because my success rate there's really low, and I'd rather be upfront than burn their cash. From $100 to $3,000 per day, how fast you test new creative drives almost all the growth. Past $3,000 per day, unit economics, bundle AOV and pre-landers set the ceiling alongside creative volume.

If you can't decide whether your next hire should run the account or write the briefs, read how to hire a Meta ads creative strategist.

The Five Checks I Run Before I Write a Single Brief

Before I write a hook, I spend five minutes in Ads Manager and the store funnel. Founders show up with a theory about why performance dipped, so I check it against the account first.

First I sort every ad by 90-day spend. I've been part of $60M+ in managed ad spend overall and I still start there, because in stalled seven-figure accounts one old ad often holds 40% to 75% of historical spend. If someone pauses it early, the damage shows up fast. On a clinical ketone brain supplement account I audited, the team paused a hero ad carrying roughly 41% of spend after a short wobble, and acquisition cost climbed from $150 to $400 over the next five weeks.

Second, I check whether the ad lost the buyer or the store did.

My benchmarks are 15% Landing Page View to Add-to-Cart, 50% to 60% Add-to-Cart to Initiate Checkout, and 50% to 55% Initiate Checkout to Purchase, with 43% to 46% still workable on higher-ticket items. On a men's herbal extract brand, blended ROAS dropped from 4.36 to 2.48. Checkout held at 43% while CPC tripled from $0.25 to $0.86, right after their top hormone-claim ad got rejected. The store still converted fine, so the hooks were the problem.

Third, I check the optimization event and how spread out the campaigns are. If budget's going to Landing Page View objectives or dozens of tiny ABO ad sets, the data's too thin to read. Fourth, I make sure decisions come from 7-day click and first-party store orders, since 1-day view numbers make weak ads look good.

Fifth, I open Account Quality and look at the active spenders. Editors sometimes swap the on-screen text after a medical claim rejection and leave the flagged voiceover in the cut. I also check how much of the spend sits on videos other sellers ran first. Ripped content should be 3 to 5% of what you run at most and heading to zero, it only works short term.

Sub-Avatars: How One Audience at 10x Frequency Becomes Four

In paid social creative strategy now, the ad does the targeting that interest stacks and lookalikes used to do. I haven't seen lookalike success in a long time. It's usually outperformed by broad, so I keep the targeting wide and let the opening frame, the spoken problem and the setting pick who stops scrolling. I cover how that works in Meta Andromeda creative strategy.

When an account hits a spend ceiling, founders assume the whole country's tapped out.

I audited a US lash extension supplier spending $3,000 per day. 30-day frequency had climbed near 10x across roughly 247,000 people, so raising the budget only paid higher CPMs to show the same ads to the same crowd. Then I looked at the ads. Every video and static spoke to veteran lash technicians working out of established studios, so one group saw ten impressions a month while the buyers next to them saw nothing made for them.

I left the campaign settings alone and split that one audience into four sub-avatars.

• Veteran lash technicians care about retention speed, reliable glue curing and shorter appointments.
• Cosmetology students want to pass practical exams without fighting unpredictable adhesive.
• Academy trainees want starter-kit value, predictable dry times and fewer client callbacks.
• Salon owners buying in bulk care about margin per tray, consistent curl quality across staff and fewer stockouts.

If you change a few caption words on the same studio footage, Meta reads it as a duplicate and it competes with its own parent ad. So each of the four gets its own brief with a different speaker, a different room and a different opening frame.

Awareness Routing: Which Ad Goes to Which Page

Ecommerce creative strategy also covers the page behind the click, because the page has to match what the buyer already knows. Someone who just found out why their legs ache at night won't buy off a discount-heavy product page. That page wasn't built to explain anything, so Add-to-Cart drops below the 10% warning line.

I saw this on a compression socks brand I audited. The whole account was bottom-funnel offer statics going straight to the product page. Buyers who already wanted compression socks converted decently, but cold prospects with sore calves bounced before reading a bullet point.

A problem-aware buyer gets a story or explainer video that names the daily symptom first, and the click goes to an advertorial or listicle that explains the cause before the product shows up. A solution-aware buyer gets a comparison or a demo that shows why your design fixes what the standard option gets wrong, and that goes to a comparison page or a detailed PDP. A product-aware buyer gets proof statics and offer statics that clear up sizing, shipping speed or price on the PDP itself.

I check where the live ads point too, because accounts split traffic by accident.

On a skincare account, the running ads went to a mix of the skin sagging collection, the single jar page and the homepage. Since the skin sagging page leads with the bundle, I moved all of them over there so more people land on the bundle.

The Product Truth Is the Angle

Exaggerated copy falls apart once real buyers touch the product or read the comments.

On a hammered cookware brand I audited, buyers were leaving angry comments on the top ads because the copy promised a "pure titanium" pan. The pan's actually a three-layer build, with titanium bonded to an aluminum core and induction steel. Pure titanium on its own heats unevenly and fails on induction stovetops. So I built the angle on the three layers, since that build heats evenly, works on induction, and weighs far less than cast iron for older cooks with wrist pain.

A protective motorcycle apparel brand selling to riders aged 45 to 80 had the opposite problem. Their product page left out the hip armor pads, and their Facebook ads creative strategy was all about breathable, waterproof denim. Older riders spend serious money on gear so they don't break bones, so I'd put the hip armor at the top of the product page and write the briefs around crash protection versus ordinary riding jeans.

Even when the product truth is clear, AI-written scripts wreck it with metaphor verbs. Someone hearing it once on a phone can't picture "locks into cellular repair" or "flips your recovery switch". Every line needs a physical picture a nine-year-old can see. Here's a made-up example for a magnesium sleep powder.

Before a script goes to your editor, I read it out loud and mark the lines that come out.

How I Read a CBO After Seven Days

I keep the testing setup simple so the auction gives me clean answers. Everything runs in consolidated Broad CBO campaigns, and spend share is the test. I don't set ad set spending minimums to prop up a weak ad, I don't move it into an ABO to force spend, and I don't usually run cost caps or bid caps.

I wait the full seven days before I grade anything. If you judge an ad on day two or three, you'll cut steady converters before delivery settles. The pacing's in how I run Meta ads creative testing and how many ads to test in a Meta CBO.

An ad under 10% of campaign spend after seven days lost.

I drop it and look at why the opening frame didn't work, because the campaign settings didn't cause it. I go through that in why Meta CBO ads stop getting spend.

Winners stay in their existing ad sets so their history doesn't reset, and every new batch goes into a new ad set in the same CBO. If an old top spender creeps above target CPA on 7-day click, I check the 14-day and 30-day windows, blended store MER and mid-funnel feeder lift before I touch it. Even if it's weak on all three, it stays live until a challenger takes majority spend on its own.

What a Week of Creative Strategy Produces

Briefs don't do anything sitting in a Notion doc.

I audited a Swiss ergonomic posture brand spending 150 CHF per day. The founder wanted to hire media buyers to set up bid strategies while fifteen approved briefs sat in draft folders, because their editor was shipping zero cuts. The bottleneck was production, so the fix was dropping the bid cap hunt and getting finished edits out every week. If your editors can't keep up, I'll slow the briefs down until they catch up.

I start with the research doc. I go through 1,000+ of your reviews, your ad and TikTok comments and 9 competitors to pull the angles in your customers' own words. I also set up the growth guide, which tracks every concept we test and what we make next.

After that it's five briefs a week with three variants each. Every brief starts from a format that's already holding spend in your account or in a nearby niche, then swaps in a new sub-avatar or visual world. The three variants change the hook and bridge on one angle, so you're testing the opening and nothing else. Each one has beat-by-beat clip direction that tells the editor where real creator footage is a must and where B-roll carries it. One brand sold out after one of our ads, DCT 38, which spent $52,048 at 2.21 ROAS with 1,276 purchases.

You can see filled-in video and static templates in how I write ad creative briefs, or see live creative strategy work across DTC categories. If your editors are ready to move, book a discovery call and I'll go through your account with you.

Questions Founders Ask Me on Upwork Calls

Should I hire a media buyer or a creative strategist first?

If you're already in Broad CBOs optimizing for Purchase and your store converts near benchmark, another media buyer adjusting bids won't move your acquisition cost much. You need more angles and more new creative going into the auction, so hire the strategist first. My rate is $600 per week for five briefs with three variants each, in 60-day blocks. Media buying only is $375 per week.

How long before new creative briefs move account CPA?

It usually takes two to three weeks from brief delivery to live data, depending on how fast your editor turns cuts around.

Once a batch is in the CBO, each ad needs seven full days before spend share tells me anything. So plan on about a month before the first batch has a real answer.

Can AI generate our ad scripts and static ads on its own?

Raw AI drafts read fine in a doc and fall flat on a phone, because they stack vague words, choppy fragments and claims you can't film. Image models can build a strong scene around your real product photo. I still rewrite every headline and body line in the words your buyers use in their reviews.

What access do you need before writing week one briefs?

I need read access to Meta Ads Manager so I can see the real history, read access to Shopify or Triple Whale for the store numbers, and your raw product photos and creator footage. Once I've got those three, I start the research doc and write the first five briefs.