Meta’s pitch is simple. Give us your website URL and your budget. We will handle the creative, the targeting, the bidding, the placements, and the spend allocation. Everything. You just read the results.
65% of all advertisers are already running campaigns through Meta’s Advantage+ suite. Those running multiple Advantage+ campaigns together have seen a 32% decrease in cost per acquisition. Click-through rates are running 11 to 15% higher than manual campaigns. CPM costs are 32% lower.
The numbers are real. The question is what they are actually measuring, and what you are quietly giving up to get them.
What Meta Advantage+ Is Actually Doing Inside Your Account
Advantage+ is not one tool. It is a suite of AI systems working together, each one taking over a different part of your campaign.
- Advantage+ Audience replaces manual targeting. Meta’s AI decides who sees your ads based on predicted conversion likelihood, not your demographic inputs.
- Advantage+ Creative generates multiple versions of your images and copy, tests them automatically, and serves the combinations it predicts will perform best.
- Advantage+ Placements decides where your ads appear across Facebook, Instagram, Messenger, and the Audience Network.
- Advantage+ Shopping Campaigns wraps all of this into a single campaign structure for ecommerce brands.
For tasks that are purely mathematical, this works well. Bid management, placement optimisation, audience expansion, budget pacing. These are problems where an AI processing millions of data points in real time will always beat a human checking in once a day.
The issue is not what the AI does well. It is what it cannot do at all.
The Variable Meta’s AI Cannot Provide: Brand Context
Meta’s AI has been trained on tens of thousands of campaigns. It knows what drives clicks. It knows what leads to conversions. It does not know your brand voice, your values, your tone, or what makes your customers loyal rather than just transactional.
When the AI generates creative on your behalf, it optimises for what has worked across all advertisers. Not what makes you different from every other advertiser in your category.
This creates what some are calling a race to the middle. Every brand using the same AI creative system starts producing ads that look and sound the same. The algorithm optimises toward the mean. Your distinctiveness, the thing your best customers actually respond to, gets averaged out.
A business that has spent years building trust through a specific tone and offer structure can lose that overnight with one round of AI-generated creative that sounds nothing like them. The click-through rate might improve. The customer lifetime value might not.
At DNM Digital, we have seen CPA drop on paper while long-term revenue suffers because the AI was optimising for the easiest conversions, not the best customers.
The Measurement Problem Nobody Is Warning You About
Meta’s reported results for Advantage+ look good because Meta is reporting them. The 32% CPA improvement is a platform-level average across all advertisers. Your account is not average.
There are three things to check before you trust the numbers in Ads Manager:
1. Is It Counting the Right Conversions?
Advantage+ optimises for conversions. But if your conversion events are set up loosely, the AI will find the easiest ones to claim. Add to cart instead of purchase. Page views instead of leads. The volume looks impressive. The revenue does not match. Your data and analytics setup needs to be clean before you hand targeting decisions to an algorithm.
2. Is It Spending on New Customers or Existing Ones?
Without proper audience exclusions, Advantage+ will serve ads to people who have already bought from you, people on your email list, and repeat visitors who would have converted anyway. The AI counts those as wins. Your acquisition cost is being understated. If you want to understand the real value of each customer, use the LTV calculator to separate acquisition performance from retention performance.
3. Is Your Budget Large Enough for the Algorithm to Learn?
Meta’s Advantage+ results were driven largely by advertisers with significant historical data and conversion volume. If your account is generating fewer than 25 to 50 conversions per week, the algorithm does not have enough signal to optimise properly. It will spend money testing while it learns. On a small budget, that testing phase is not cheap.
What to Hand Over and What to Keep
The most effective approach is selective automation, not complete automation. Here is the dividing line:
- Hand over: bid management, placement decisions, audience expansion, budget pacing between ad sets. The AI is genuinely better at these than any human.
- Keep control of: creative direction, brand tone, offer strategy, and landing page experience. These require context the algorithm does not have.
In practice, this means you write the headlines. You set the creative brief. You decide what the offer is and how it is framed. Then you let Advantage+ distribute and test those assets across placements and audiences.
If you hand over the creative as well, you are not using AI to amplify your brand. You are outsourcing your brand to an algorithm that has never heard of you. And if your conversion rate is not where it needs to be, no amount of AI optimisation will fix what happens after the click.
Talk to Us About Your Meta Ads Setup
How to Test Whether Advantage+ Is Actually Working for You
Before you switch your entire account to full automation, run a proper comparison. Set up a duplicate campaign with your current manual settings alongside an Advantage+ version. Run both for 30 days with equal budget. Then compare cost per acquisition, customer quality, and repeat purchase rate. Not just click-through rate and CPM.
Cheaper clicks are not the goal. Revenue is. The brands that will get the most from Meta’s AI are the ones that treat it as a distribution and optimisation layer, not a replacement for strategy. Give the algorithm a defined lane: strong creative inputs, tight audience exclusions, clean conversion tracking. Then let it do what it does better than any human.
The Bottom Line
Meta’s AI advertising tools are genuinely powerful. The performance data is real. And for the operational parts of running a campaign, automation will outperform manual management almost every time.
But the brands that hand over everything (creative, tone, offer, measurement) are the ones that end up with lower CPAs and weaker businesses. The algorithm will find you conversions. It cannot build you a brand.
If you want to know whether your current Meta Ads setup is working for your business or just for Meta’s reporting, we are happy to take a look.
