Meta Advantage+ is the default now. Here is your move

Meta is no longer asking whether you want its AI to run your campaign. As of February 2026 it merged the manual and Advantage+ build flows into one, and the automated path is now the default for every new Sales, Leads and App Promotion campaign. That means targeting, placements, budget split and most creative tweaks are pre-selected for you the moment you start. If you run your own ads, the lever you keep is creative and clean conversion data. The dials you used to obsess over are mostly gone.

What happened

Two changes landed close together and they compound.

First, the rename. In 2025 Meta retired the "Advantage+ Shopping" name and called it Advantage+ Sales. The scope grew with the name. It now covers e-commerce sales, lead generation and app installs, not just product catalogs. So when people say "Advantage+" in 2026 they usually mean this campaign experience, not one shopping product.

Second, and bigger for small advertisers: in February 2026 Meta collapsed the old "Manual" versus "Advantage+" choice into a single campaign setup. You no longer pick a lane at the start. You get one flow with the AI options already switched on. New campaigns launch with Advantage+ Audience, Advantage+ Placements, Advantage+ creative enhancements and campaign budget optimization pre-selected. Meta confirms this on its own Advantage+ campaign help page.

Underneath that, the manual controls have been thinning out for a while. Detailed-targeting exclusions were removed in early 2025. Several creative enhancements now run by default. And the most recent shift worth knowing about: Meta added a control in March 2026 that lets you cap the share of budget going to existing customers, an admission that broad automation was over-spending on people who already buy from you.

Why it matters

For a big account with a media team, this is a workflow change. For a small advertiser running their own ads, it shifts where the work is.

Three things move out of your hands by default.

Targeting becomes a suggestion, not a setting. With Advantage+ Audience, only your location and minimum age are treated as hard limits. Everything else, your interests, your custom audiences, even your lookalikes, are "audience suggestions." Meta uses them as a starting hint and then spends outside them whenever its model thinks it will perform better. Jon Loomer's breakdown of Advantage+ Audience versus original audiences walks through exactly how soft those inputs are. If your product only suits a narrow buyer, that looseness can quietly waste spend.

Placements get handed over too, and the budget can drift. For Sales and Leads campaigns, when you exclude a placement at the ad set level, Meta now shows a checkbox (frequently on by default) that lets it spend up to 5% of the ad set budget on each excluded placement when it predicts better results. Exclude four placements and up to 20% of your budget can land on surfaces you tried to block. PPC Land documented this 5% spend on excluded placements behavior. The practical risk for a smaller budget: money trickling onto cheap, low-intent inventory you specifically did not want.

Creative becomes the main lever. When the system owns who sees the ad, where, and for how much, the inputs you still genuinely control are the creative and the offer. That is not a slogan, it is just what is left on the dashboard. A default Advantage+ campaign with one tired ad gives the AI nothing to optimize between. The same campaign with five honestly different angles gives it room to find a winner. This is why we keep saying the work has moved upstream into the creative and the funnel. The same logic applies to where those leads land after the click, which is why a tight follow-up system like the one in our GoHighLevel setup for local businesses matters more, not less, once Meta owns the front end.

Your move

Stop fighting the automation and start steering it. Five things to do this week.

Feed it strong, varied creative. Give every campaign at least four to six genuinely different ads, not four crops of the same image. Different hooks, different formats, different angles. The model can only test what you give it.

Set the exclusions that actually matter. You lost interest exclusions, but you can still exclude existing customers, recent purchasers and audiences you never want to pay to reach again. Those still hold. Upload them and use them.

Lock down placements you do not trust. If brand safety or junk traffic is a concern, do not rely on the ad-set toggle. Set placement exclusions at the account level (Advertising Settings, then Placement Controls). Account-level exclusions override everything and have no 5% leak checkbox. The Optimizer's note on blocking placements in 2026 covers the gotchas.

Watch for budget drift. Check the placement breakdown weekly. If spend is creeping onto Audience Network or surfaces that never convert, that is the 5% leak at work. Uncheck the limited-spend box at the ad set level, or move the exclusion up to account level.

Keep your conversion signal clean. All of this AI runs on the data you send back. A broken pixel, a missing Conversions API event, or sloppy event setup and the system optimizes toward noise. Make sure your purchase or lead events fire correctly before you blame the targeting.

Tools and resources

You do not need new ad software for this. You need the basics working: clean tracking, fast follow-up, and a place for leads to land that does not leak.

If lead gen is the objective, the weak point is almost never Meta. It is what happens after the form. Leads sit, nobody replies for hours, and the spend is wasted no matter how good the targeting was. A connected CRM that fires automated SMS and email the second a lead comes in fixes more wasted budget than any audience tweak. We use GoHighLevel for exactly that, and our full GoHighLevel review covers where it is strong and where it annoys us. If cost is the question first, the GoHighLevel pricing breakdown has the real bill including usage fees.

For the wider picture of what a lean marketing setup looks like in 2026, our marketing tech stack guide lays out the tools worth paying for and the ones you can skip.

Worth watching

A few things we are keeping an eye on.

Whether the manual escape hatches survive. Each release has quietly removed another control. If audience or placement exclusions get softened further, the only real lever left is creative and offer, so build that muscle now.

The existing-customer budget cap. The fact that Meta shipped a control to stop over-spending on current customers tells you it was a real problem. If you sell repeat-purchase products, set that cap and watch whether new-customer volume actually improves.

Creative production becoming the bottleneck. As targeting flattens out, the advertisers who win are the ones who can ship more honest creative variations, faster. That is a content and production problem now, not a targeting one. Worth thinking about how you produce ads, not just how you target them.

More reads on what is changing across the ad platforms live under our ad platforms coverage, and the full article archive has the rest. If you want our take on a specific tool in this setup, the GoHighLevel breakdown is a good place to start.