Meta dedicated the last two years to dismantling the campaign machinery that advertisers spent over a decade perfecting. Segmented ad sets, manual audience controls, tightly gated targeting: much of the hand-setting that marketers used to do themselves is either deprecated or on its way out. The platform is betting big on AI, and it wants fewer inputs from you and more creative and more conversion data for its models to run on.

Those changes might feel like a loss of control. But when we attended Meta’s Performance Marketing Summit, we learned that Meta is providing more clarity in the areas where advertisers have often been flying blind. What you give up in structure, you regain in measurement, creator distribution, and catalog intelligence. Here are some ways to put that newfound transparency to work in your campaigns.

Broaden Your Ad Goals Beyond Revenue

“Test and learn” has always carried a tension for ecommerce brands operating against hard revenue targets. Every experiment spends a portion of your budget on an unproven approach, so you’re risking the number that you promised leadership you’d deliver. When the platform underneath your campaigns keeps shifting, your instinct is to revert to tried-and-true settings and prioritize hitting this quarter’s number over compounding gains later on.

But that instinct can get expensive. Brands that refuse to adopt new settings to preserve short-term ROAS are optimizing against a version of Meta that will soon disappear. The cost is a slow, invisible decay in performance as the platform rewards behaviors you’ve stopped exhibiting. If your revenue targets are strict, analyze true contribution rather than vanity metrics. A Meta rep is required for custom attribution modeling, so involve one early.

Separate Correlation And Causation

An ad that gets credit for a transaction because it sat somewhere in the click path tells you almost nothing about whether it drove the purchase or intercepted a customer who was already going to buy from the brand. This is what incrementality measurement solves.

Incrementality answers the question that attribution can’t: Would this conversion have happened anyway? Meta’s conversion lift studies are free. Use them. But remember that a test result has a shelf life. The platform changes. Seasonality shifts. Creative fatigues. Treating a Q1 read as gospel in Q4 is how brands convince themselves they’re data-driven while drifting off course.

Think Of Creators As Distribution Channels

Your prospecting campaigns are grinding against people who’ve learned to ignore branded ads. Creators reach audiences actively open to new products. That insight helped influence Meta to rebuild algorithms that recommend creators who fit your product and brand.

Soon creators will be able to tag products for purchase inside Reels, collapsing the distance between attention and transaction. The expansion of creator affiliate partnerships into Amazon in the US and Mercado Libre in LATAM signals that creator content is becoming a shoppable surface rather than an awareness afterthought. 

Partnership ads run content in a high-performing format that you can measure. Plus, they’re available at a time when many advertisers struggle to diversify their creative assets at the volume that Meta’s new best practices require. Scaling up from 3–5 ads to 10–20 ads per audience is a spike in workload for creative teams. Running partnership ads can help brands align their campaigns with the volume and diversification needed to stay competitive on Meta without straining their creative teams.

Put Your Catalog To Work

Catalogs are expanding beyond retail into streaming, gaming, publishing, and other industries. But two catalog capabilities within the Meta suite still go underused:

Catalog Product Video

Almost no one runs product video, so the brands that do stand out in a feed of static tiles. Upload video assets to your catalog while your competitors rely on getting consumers’ attention with standard product images against a white background.

Business AI

Meta’s Business AI agent analyzes your site and catalog in real time to answer customer questions and build trust at the exact moment buyers are weighing their options. The feature is new and still in development. Another feature that Mindgruve’s excited about: AI-assisted clicks and AI-assisted conversions, two metrics that track impact directly. For the first time, you’ll see how the conversational layer contributes to a sale.

Start With This Checklist

Act on all these recommendations with a few next steps:

Measure Incrementality

Treat incrementality as a team sport. Pull analytics, finance, and leadership into the same room before you run any tests. (An incrementality finding that contradicts the last-click dashboard will get killed on arrival unless other departments understand why the two numbers disagree.) Feed the results into whichever measurement model you trust the most.

Optimize Catalog Intelligence

You don’t need to promote your entire catalog to see results. Start with the top sellers, prove the mechanics, then keep going. Consolidate to a single catalog. Fragmentation across multiple catalogs wears away at the signal that Meta’s models need to show your brand to the right shoppers.

Plan For Creator Distribution

Match the specialist to the objective before you match the follower count to the budget. Followers fill feeds. Creators fill carts. Take time to cast for the job.

New Controls Mean New Possibilities

The old structural levers are going away. Our advice: stop protecting the best practices you’re in the process of losing. Redirect that energy toward the measurement, creator, and catalog capabilities that Meta is actively expanding right now. That much, at least, is within your control.