How Does Facebook Evaluate Business Pages? (2026)

Facebook is scoring your business page right now. Quietly, constantly, on trust and quality and how your customers actually feel, across a set of internal signals you will never lay eyes on. Those scores decide what your ads cost, how far they reach, and how high you can scale. Knowing what Meta is really looking at is the line between fixing the real problem and guessing at it.

Most advertisers only ever see the surface. Ads Manager, campaigns, the occasional policy notice. Underneath all of it sits an evaluation layer that has been running the whole time.

How does Facebook evaluate a business page?

Think of Meta’s platform as two layers.

The one you use sits on top. Ads Manager, campaigns, creatives. Below it runs an internal evaluation layer that quietly scores every entity attached to your advertising: your page, your ad account, your Business Manager, your business as a whole. In our experience those scores are calculated continuously, updated the moment new data lands, and never shown to you. You can build a technically flawless campaign and still get throttled, because the layer underneath read your business as low trust or low quality.

Here is the part that trips people up. Facebook isn’t only judging your ads. It’s judging you, and pricing your delivery to match.

The signals Facebook is actually watching

Based on what we see across the accounts we manage, the whole evaluation rests on a handful of internal signals that stack into a hierarchy.

• Business Manager reputation. The top of the stack. It sets a ceiling, and individual assets rarely climb higher than the BM’s standing allows.
• Ad account trust level. Built from compliance history, payment reliability, and how the account behaves over time. It shapes approval speed, how much you can run while in review, and how much you’re allowed to spend.
• Your Facebook feedback score. The 0 to 5 signal that comes straight from post-purchase customer surveys. Invisible since late 2024, but still live and still shaping delivery and cost.
• Asset-level delivery quality. An internal rating of how well each asset, page or account, competes and delivers. A weak one means a higher CPM and less reach at the very same bid. We get into this, the “HIVA” concept, in the feedback score guide.

None of these sit in isolation. A weak feedback score drags down trust and delivery quality. A low-reputation Business Manager caps everything below it. Fix one signal on its own and you barely move the needle, because the system reads you as a whole.

What Facebook actually measures

These scores aren’t magic. They’re assembled from things you do, or don’t do.

• Customer experience. The single biggest input. Post-purchase surveys ask your buyers about product quality, shipping speed, and service, and that data feeds the feedback score directly.
• Public sentiment. Your reviews and recommendations, plus engagement and how responsive you are, feed the wider trust picture. That’s the broader subject of your page reputation and credibility.
• Compliance and history. Ad violation rate, payment reliability, and how the account has behaved over time.
• Legitimacy signals. Verification, a complete profile, coherent branding, consistent activity. The markers of a real business rather than a disposable one.

How the evaluation shows up in your results

You never see the scores, so you feel them as symptoms instead.

A favourable evaluation buys you cheaper reach, smoother delivery, faster approvals, and a higher ceiling to scale into. A degraded one does the opposite. A higher CPM floor, throttled or stalled delivery, more rejections, a wall you keep hitting at a certain spend, and at the extreme, real ban risk. When several of those land at once and your campaigns haven’t changed, the evaluation layer is almost always where to look.

What you can and can’t see in 2026

You can see some of the downstream indicators. The Account Quality dashboard with its violations, restrictions and status, and the ad-level quality, engagement and conversion rankings. These are endpoints, not early warnings.

You can’t see the scores that actually drive the system. Not the feedback score, pulled from view in late 2024. Not the account trust level, not the Business Manager reputation, not the internal delivery-quality rating. That gap between what gets measured and what gets shown is exactly why so many advertisers mistake an account-level problem for a campaign one.

How to be evaluated favourably

You can’t game an invisible system. But you can feed it well.

The levers that matter most are the ones already in your hands. Keep the customer experience tight, which means shipping, product and support. Make your happy customers visible in your reviews. Stay compliant and verified. Keep your account structure clean. In practice that’s the exact same work as improving your feedback score and building page credibility, because the evaluation and the experience are, in the end, the same thing.

Want to know how Meta is really evaluating you?

Because the scores stay hidden, working out where your business actually stands means reading the signals that are visible and correlating them with your operations. That’s what our team does every day. Across 1,000+ e-commerce accounts, Unlimited Scaling audits how Meta is evaluating your assets, pinpoints what’s dragging you down, and fixes it at the root. No guarantees, no shortcuts. Just a clear read on where you stand and how to move up.

Every page is evaluated on its own

Here is a detail that reframes everything above. In our experience Meta evaluates each page as its own separate entity, with its own trust standing, its own feedback score, and its own review history. It isn’t one verdict on your brand that trails you everywhere you go. It’s a judgement made asset by asset. And that has two consequences most advertisers miss.

• Bad signals don’t stay contained by luck. They stay contained by structure. A cluster of complaints or a poor recommendation rating drags down the page it lands on. From what we’ve seen, it does not automatically contaminate a separate page selling the same product, as long as the two aren’t tightly linked.
• One page is one point of failure. If everything you run sits behind a single page, and that page gets restricted in a verification or flag wave, 100% of your delivery can stop overnight. The brands we work with that run the same offer across several avatar-specific pages treat each page as insurance. A bad week on one page costs a fraction of revenue, not all of it. We go deeper in our multi-page strategy guide and on backup pages.

There’s a catch worth flagging. In our experience Meta actively tracks association between assets. If a backup page is linked to a Business Manager that’s already restricted, it tends to inherit the flag rather than escape it. This isn’t Meta-published, but it’s why the isolation only holds when pages sit on genuinely separate profiles and business managers, sometimes with an agency ad account adding a layer of separation.

Andromeda now scores page-to-audience fit

The evaluation isn’t only about trust and compliance anymore. With Meta’s Andromeda ranking update, the system also appears to weigh how well a page’s identity matches the person it’s shown to. In our reading of how delivery has shifted, Andromeda matches the page name, positioning and hook to a specific segment. So a coherent, specific page, think “dermatologist specialist” or “garden care community,” gets targeted far more precisely than a generic brand page trying to speak to everyone at once.

This is why the same ad copy can perform completely differently on two pages. It’s the system matching message to audience, not luck. A vague page hands Andromeda a weak signal to work with. A specific one lets it find the right buyer at a lower cost.

Andromeda also folded page reviews and recommendations more firmly into the picture. Facebook asks users whether they recommend a page and calculates a rating from 0 to 100%. In our experience that recommendation rating now feeds ad performance directly, and it works best paired with a strong post-purchase feedback score. Lift one without the other and you’ll see little. The two move the needle together.

Why a brand-new page starts at a disadvantage

Page history is a signal in its own right. From what we’ve seen, Meta treats a brand-new page with no content, no followers and no activity as high risk by default. And high risk tends to mean a higher CPM, weaker delivery, and a faster path to restriction the moment it starts spending.

The fix isn’t a trick. It’s making the page look like a real business before you scale spend behind it. In practice that means posting a little, gathering some genuine organic engagement, and letting the page build a short track record, so a real person browsing it would believe it. In our experience a warmed page can outperform a page that gets flagged on day one by roughly 10x. The evaluation rewards patience here precisely because history is one of the few legitimacy signals a disposable page can’t fake overnight.

FAQ

---

Written by Mouss, founder of Unlimited Scaling, an agency that has helped 1,000+ e-commerce brands recover and protect their Meta ad assets. Based in Bali, he has spent 8+ years inside the mechanics of Meta’s ad ecosystem, feedback scores, HIVA tiers, bans and appeals, and shares field data from real client cases. Follow him on Instagram @mouss_unlimitedscaling.