Meta made $134 billion in advertising revenue in 2023. The vast majority of it came not from sophisticated brands with dedicated media agencies, but from small and medium businesses who largely have no idea whether their ads are working. That’s not an accident. It’s the product.
This isn’t a “social media bad” screed. Meta ads, used properly, can generate real returns. But the way Meta actively nudges small businesses to spend money — the notifications, the Boost button, the dashboard that shows you exactly the numbers it wants you to see — is worth understanding before you hand over your card details.
The Boost Button Is a Trap
You post something on your Facebook business page. A few hours later, a notification appears: “Your post is performing better than usual. Boost it to reach more people.”
This notification is not designed to help you. It is designed to get you to spend money with as little friction as possible, before you’ve had a chance to think about what you’re trying to achieve.
Boosting a post is the worst way to advertise on Meta. It gives you almost no targeting control, no conversion tracking, no meaningful optimisation objective, and no ability to properly measure results. What it gives you is a number that goes up — reach, impressions, “people reached” — which feels like success but tells you absolutely nothing about whether anyone bought anything, called you, or did anything other than have your post briefly appear in their feed.
The Boost button exists at the post level rather than inside Ads Manager for one reason: it’s designed to catch people who don’t know what they’re doing. Anyone who knows how Meta advertising works uses Ads Manager. The Boost button is for everyone else, and Meta has made it as easy and impulsive as possible to click.
The Organic Reach Squeeze
Here’s something Meta would rather you didn’t think too hard about. In 2012, when you posted on your Facebook business page, roughly 16% of your followers saw it organically. By 2014 that was down to 6%. Today it’s somewhere between 1% and 5% for most pages, and trending downward.
This wasn’t a technical limitation. It was a deliberate strategic decision. By throttling organic reach, Meta created a problem — your followers can’t see your content — and then sold you the solution: pay to reach the audience you already built.
The playbook worked extraordinarily well. Businesses had invested years building Facebook followings. When reach collapsed, many felt they had no choice but to pay to access their own audiences. The alternative — abandoning the platform — felt too costly.
This is worth knowing not because you should never advertise on Meta, but because you should go in with clear eyes about who this relationship is set up to benefit.
How Meta Lies About Conversions
This is the one that genuinely costs businesses serious money, and it’s the least understood.
When Meta tells you that your ad generated 47 conversions, what does that actually mean? By default, Meta attributes a conversion to your ad if someone saw or clicked the ad and then converted within a certain window. The default attribution window is 7-day click or 1-day view.
That second part — “1-day view” — is where the sleight of hand happens.
A view attribution means: someone scrolled past your ad without clicking it, and then within 24 hours they went to your website and bought something. Meta counts that as a conversion driven by your ad.
Did your ad cause that conversion? Maybe. Or maybe that person was already planning to buy from you, Googled your brand, and bought — and Meta just happened to have shown them an ad at some point that day. You’ll never know, and Meta’s dashboard won’t tell you.
It gets worse. Meta’s pixel fires on your website. Meta’s attribution model is built by Meta. Meta is the one counting conversions and reporting them back to you. There is an enormous and obvious conflict of interest in letting the platform that sells you the ads also be the sole arbiter of whether those ads worked.
Independent studies that compare Meta’s reported conversions against server-side data or third-party attribution tools consistently find that Meta overcounts conversions — sometimes by 50%, sometimes by several hundred percent. The gap between what Meta claims and what actually happened varies by industry, audience, and product, but it almost always favours Meta.
Estimated Reach Is Meaningless
When you set up an ad campaign, Meta shows you an estimated daily reach: “Estimated audience size: 180,000–520,000 people.” This number gives the impression of precision and scale. It is neither.
The range is enormous by design — a factor of nearly three in the example above. It’s based on historical averages across similar campaigns and tells you very little about what will actually happen with your specific ad, your specific creative, and your specific budget. It’s a number that makes you feel like you’re making an informed decision when you’re not.
Similarly, the cost estimates Meta shows you before you run a campaign are essentially fiction. Actual costs vary wildly based on your creative quality, your audience, your objective, what everyone else is bidding at that moment, and a dozen other factors Meta doesn’t disclose.
Advantage+ and the Removal of Control
Over the past few years, Meta has been systematically removing advertiser control under the banner of “AI optimisation.” Products like Advantage+ Audience, Advantage+ Placements, and Advantage+ Shopping Campaigns automate targeting, placement, and creative decisions that advertisers used to make themselves.
The pitch is that Meta’s AI knows better than you who to show your ads to. That may sometimes be true. But there’s another reason Meta wants you to use Advantage+: when the algorithm controls the targeting, you can’t see exactly who you’re reaching, can’t exclude audiences that don’t make sense, and can’t learn the specific segments that respond to your offer. You become more dependent on Meta and less capable of running ads yourself.
Advantage+ campaigns also tend to spend budget faster and more broadly than manually controlled campaigns. Faster spend = more revenue for Meta in the short term. Whether it’s better for you depends entirely on your specific situation — but Meta’s defaults are not calibrated in your favour.
What Actually Works on Meta (If You’re Going to Use It)
None of this means Meta ads are useless. It means you need to use them on your terms, not theirs.
- Never boost posts. Always use Ads Manager. Set a proper campaign objective tied to something that actually matters — conversions, leads, or at minimum, link clicks to a specific landing page.
- Set up server-side tracking. The Meta Conversions API sends conversion data directly from your server rather than relying solely on the browser pixel. It’s more accurate, more privacy-compliant, and harder for Meta to inflate.
- Use your own attribution. Compare Meta’s reported conversions against your CRM, your Google Analytics, your actual orders. If Meta claims 50 conversions and your CRM shows 12 new customers, that gap is telling you something important.
- Change the attribution window. Switch your reporting from the default “7-day click, 1-day view” to “7-day click” only. You’ll see fewer conversions reported — but they’ll be more honest.
- Retargeting over cold audiences. Meta’s strongest use case for most small businesses is retargeting: showing ads to people who’ve already visited your website. The intent signal is real, the audience is warm, and the attribution is less muddied. Start there before spending on cold audiences.
- Test creative aggressively. The single biggest variable in Meta ad performance is the creative — the image or video. Not the targeting, not the audience, not the budget. Make multiple versions and let them compete.
The Bottom Line
Meta is a business. Its product is attention, and its customers are advertisers. The small business owner who boosts a post on impulse after a notification is not a priority customer — they’re a revenue stream that requires almost no effort to maintain. The dashboard is designed to show them enough positive-looking numbers that they keep spending without asking hard questions.
If you use Meta ads with proper tracking, real objectives, and independent verification of results, they can absolutely work. But go in knowing that the default settings, the default attribution, and the default recommendations are optimised for Meta’s revenue, not yours.
