If your Meta ads used to work and now they don't, the temptation is to spend more. Raise the budget, let the algorithm learn, wait for the ROAS to come back.

It doesn't come back.

The problem isn't your budget. Blaming the algorithm is almost always the wrong diagnosis. Meta's machine learning has got better, not worse. Advantage+ campaigns are genuinely outperforming manual setups for a lot of DTC brands right now. If your results are declining, something else is broken.

After running paid media for fashion, beauty, and lifestyle brands across Meta, Google, and TikTok, we've narrowed it down. Poor performance almost always comes from one of five causes. Get the diagnosis right, and the fix is usually straightforward.

Meta CPMs have gone up. That's the baseline.

Before the diagnosis: Meta advertising costs have risen structurally, not cyclically. CPMs are up 25-40% across most DTC categories compared to two years ago. iOS privacy changes reduced signal quality. More brands are bidding for the same eyeballs.

If your results are worse than they were in 2022, some of that is the new environment. You can't outspend your way back to 2022 economics. What you can do is get more efficient, which means fixing the things within your control.

Here's where to look.

The most common cause: your creative is worn out

This is the biggest one by some distance. Meta ads are fundamentally a creative game, and most brands run far too few variations.

The data is consistent: brands producing 30 or more creatives per month maintain stable CPAs at scale. Brands running three to five creatives per month see fatigue within weeks. The algorithm finds your best-performing ad quickly. Once it runs out of fresh options, performance degrades. You'll notice it as CPMs rising while CTR drops.

Signs this is your problem:

• CTR was strong six to eight weeks ago and has since dropped with no other changes
• One or two ads are carrying all the spend
• You haven't tested a new hook, format, or angle in the past month

The fix is not just making more ads. It's testing more angles. Creative testing follows a hierarchy: concept and angle first, then hook, then visual format, then body copy, then CTA. Most brands test the wrong things, like visual aesthetics and caption length, instead of the highest-leverage variables.

One more thing worth knowing: UGC and creator-style content reduces CPA by 23% on average versus brand-produced creative. The polish that makes your brand look premium in editorial content often works against you in the feed. Real footage, real people, a slightly imperfect talking-head cut — these earn trust in a way studio shots don't.

You're optimising for the wrong event

Meta needs signal to optimise. The campaign objective tells it what signal to chase. If that signal doesn't match your actual goal, you get cheap results that don't turn into revenue.

A common mistake: running a traffic or engagement campaign and wondering why the ROAS is poor. Traffic campaigns optimise for clicks. Engagement campaigns optimise for reactions. Neither is a purchase.

For DTC brands, the default should be Purchases as your conversion event, but only if you're generating at least 50 conversion events per week, per ad set. Below that threshold, Meta doesn't have enough data to optimise properly. If you're early-stage or selling a high-ticket product, optimise for Add to Cart or Initiate Checkout first, then graduate to Purchase once volume allows.

Signs this is your problem:

• Good click-through rates, poor conversion rates
• High link clicks, low purchases
• Your campaign reports look healthy but Shopify revenue tells a different story

Your audiences are competing against each other

Audience overlap is invisible until you look for it. If you're running multiple ad sets targeting overlapping audiences, Meta's algorithm pits them against each other in the auction. You bid against yourself. CPMs rise across all sets and performance drops everywhere.

The other audience mistake: not excluding existing customers from acquisition campaigns. You're spending acquisition budget to reach people who already bought from you. Wasteful, and often annoying to the customer.

Signs this is your problem:

• Multiple ad sets with similar interest targeting running at the same time
• Broad and detailed targeting running in parallel without exclusions
• New customer acquisition costs keep rising while retention looks fine

Fix: use Audience Overlap in Ads Manager to check for conflict. Consolidate into fewer, broader ad sets. Exclude existing customers using a custom audience built from your customer list or Klaviyo data.

The ad and the landing page aren't connected

The click is only the beginning. If your ad promises a specific product, collection, or offer, and the landing page doesn't match, the conversion rate suffers.

This is continuity failure. The viewer builds an expectation from the ad. The landing page breaks it. The friction doesn't show in your ad metrics. It shows up as high bounce rates and low add-to-cart rates on the receiving page.

Signs this is your problem:

• Click-through rate looks healthy but your landing page conversion rate is below 2%
• You're sending all traffic to your homepage rather than a specific product or collection page
• The ad references an offer that's hard to find once you land

Fix: match the visual language, specific product, and offer directly. If the ad shows a bomber jacket, the link goes to the bomber jacket page. If the ad mentions 20% off, the page confirms 20% off above the fold. Remove the steps between the click and the decision.

You're reading the attribution wrong

Meta claims more credit than it deserves. The default 7-day click, 1-day view attribution window means that if someone sees your ad and buys within a week, Meta counts it, regardless of whether the ad actually influenced the decision.

Many DTC brands look at Meta's reported ROAS, see reasonable numbers, and assume things are fine. Then they look at Shopify revenue and the numbers don't match. That gap is the attribution problem.

Signs this is your problem:

• Meta ROAS looks solid but overall revenue and new customer acquisition aren't growing
• You pause Meta spend for a week and revenue barely moves

This isn't a reason to stop running Meta ads. Even with attribution inflation, Meta drives real discovery and demand for DTC brands. But it means you need a secondary view: blended CAC across all channels, alongside incrementality testing to understand the real lift your Meta spend is generating.

How to diagnose which problem you have

Don't try to fix all five at once. Isolate the variable.

Check creative fatigue first. How many active creatives are running? When did you last launch a new angle? If it's been more than four weeks, start there.

Check your conversion events. Is your campaign optimising for Purchases? Are you getting enough conversion volume per ad set per week?

Check audience overlap. Run the Audience Overlap tool in Ads Manager. If your main ad sets overlap by more than 20%, consolidate before doing anything else.

Check landing page continuity. Click your own ads as a new customer would. Does the page match the ad?

Check attribution discrepancy. Compare Meta's reported purchases with Shopify's last-click attribution. A gap of 20-30% is normal. Anything above that warrants a closer look.

What good looks like

When Meta works properly for a DTC brand, the setup is usually simple: 30-plus creatives cycling through per month, weighted toward UGC and creator-style formats. One to three broad ad sets per objective, exclusions in place, Advantage+ handling placement. The algorithm does the optimising. The brand's job is to keep feeding it new creative.

Most underperformance isn't a Meta problem. It's a creative volume problem, an audience setup problem, or a landing page mismatch. The algorithm is doing what it's told. The question is whether you're telling it the right thing.

To understand how Advantage+ fits into this — and whether it's the right move for your account — read Meta Advantage+ for DTC Fashion Brands: The 2026 Setup.

If you want a second set of eyes on your account, that's what we do. Book a discovery call.