Meta Ads in 2026: What's Working, What's Dead, and What to Do Next
If you've been running Meta ads for any length of time, you already know the feeling: something that worked beautifully six months ago suddenly stops performing, and you spend the next few weeks trying to figure out why.
Meta's ad ecosystem evolves constantly — the algorithm changes, the creative formats shift, privacy regulations reshape what data you can actually use, and your competitors are running more ads than ever. Keeping up isn't optional anymore. It's table stakes.
I work with ecommerce brands and growth-stage companies on their paid media strategy, and Meta ads are almost always part of the conversation. Here's my honest read on what's working right now, what I've stopped recommending, and what I think the next 12 months look like.
What's Actually Working in 2026
Broad Targeting (Yes, Really)
This one still surprises people, but the data keeps confirming it: broad targeting — or what Meta now calls "Advantage+ Audience" — is outperforming narrow, interest-based audiences for many advertisers.
This is a direct result of how much Meta's AI has improved. The algorithm is genuinely good at finding the right people now, often better than we are at manually defining who those people are. The days of stacking 15 interest layers and congratulating yourself on your laser-targeted audience are largely over.
What does this mean practically? Give Meta the creative, give it a budget, let the audience targeting stay wide, and let the algorithm learn. Test with strong creative rather than trying to engineer the audience.
The caveat: Broad targeting works best when you have enough conversion data for the algorithm to learn from. If you're a newer advertiser or running a brand with limited purchase history, you may still benefit from some audience signals to start.
Video Creative — But Specifically This Kind
Not all video is created equal right now. What's performing is short-form, native-feeling content that fits how people actually use the platform. Think lo-fi over produced. Think the way a creator would film something, not the way a brand would.
The content that stands out right now is: fast hooks in the first two seconds, authentic product demonstrations, creator-style testimonials, and content that doesn't immediately scream "this is an ad."
Polished 30-second brand films? Unless they're genuinely exceptional, they're getting skipped.
Advantage+ Shopping Campaigns
For ecommerce brands specifically, Advantage+ Shopping Campaigns (ASC) have become one of the best-performing formats available. Meta handles much of the heavy lifting — audience, placement, and optimization — while you focus on feeding it high-quality creative.
ASC works best when paired with a strong product catalogue, solid creative variety, and enough daily budget for the system to learn (typically $100+/day to start seeing meaningful data).
Retargeting With Intention
Retargeting isn't dead — but it's different than it was. With signal loss from iOS changes limiting how precisely you can retarget based on website behaviour, retargeting pools have shrunk and the data is less reliable.
What still works well: retargeting your email list and past purchasers (first-party data is king), video view retargeting (people who watched 50%+ of a video are warm), and engagement-based audiences.
What's worth pulling back from: site visitor retargeting with very short windows, which has become less accurate and more expensive.
What I've Stopped Recommending
Detailed Interest Targeting as a Primary Strategy
As I mentioned above: Meta's AI is better at finding your customer than you are at describing them with interest categories. Narrow interest targeting often restricts the algorithm unnecessarily and drives CPMs up.
I'm not saying never use it — I'm saying stop treating it as the foundation of your campaign strategy.
Optimizing for Link Clicks
If you're still running campaigns optimized for link clicks because it's cheaper than conversion campaigns — stop. You are paying to send curious people to your website, not buyers. The algorithm optimizes for what you tell it to optimize for. If you ask for clicks, you get clicks. You want purchases, customers, or at minimum, add-to-carts.
The only time I use link click optimization is for very top-of-funnel awareness plays where I genuinely don't care about conversion data and just want reach.
Changing Campaigns Too Quickly
This one is less about a format and more about behaviour. One of the most common mistakes I see is advertisers jumping in to make changes to campaigns before the algorithm has had enough time to learn.
Meta's ad system needs data to optimize. If you're resetting campaigns every 3-4 days because the ROAS isn't where you want it, you're working against yourself. Unless something is dramatically off, campaigns need at least 7 days and 50 optimization events to exit the learning phase. Patience is part of the strategy.
Over-Relying on Platform-Reported ROAS
Meta's reported ROAS is not the complete picture. It never was, but post-iOS 14, it's become even less reliable. Platforms measure what they can see — and with signal loss, they can't see everything.
Cross-reference your Meta-reported numbers with actual revenue in your store, first-party data, and post-purchase surveys. If Meta says you did $10K in attributed revenue but your Shopify dashboard shows a different story, the truth is probably somewhere in the middle.
What the Next 12 Months Look Like
A few things I'm watching closely:
AI-generated creative will become table stakes — but differentiation will be in the human layer. As more brands use AI tools to produce creative at scale, the volume of ads running on Meta will continue to rise. The winners will be brands that use AI for production efficiency while keeping the creative strategy and brand voice distinctly human.
First-party data becomes a real competitive moat. As signal loss continues to affect targeting accuracy, brands with large, engaged email lists and robust CRM data will have a significant advantage. Start building and cleaning your list now if you haven't.
Lead generation will see continued investment. With ecommerce margins under pressure, more brands are experimenting with lead gen as a way to move people into a lower-cost nurture sequence before converting. Expect more brand-to-DTC overlap in how companies think about their Meta strategy.
The creative testing flywheel separates winners from everyone else. The brands that will win on Meta in the next year are the ones running structured creative tests consistently — not one test a month, but ongoing, systematic creative experimentation. If you don't have a creative testing process, that's where I'd start.
The Bottom Line
Meta ads are still one of the highest-leverage paid channels available, especially for ecommerce and consumer brands. But the playbook has changed, and it keeps changing.
The advertisers who are winning right now are the ones who understand that Meta's AI is a tool to work with, not against — and who are investing in creative quality, first-party data, and strategic patience instead of trying to outsmart the algorithm with clever audience hacks.
Stop fighting the machine. Feed it what it needs, and focus your energy on the things it can't do for you: great creative, smart strategy, and a product worth selling.
