How to solve constant
account flags on Meta
(the real reason it keeps happening)
Most advertisers treat the symptom. We explain the root cause of repeat Meta flags and the system-level fix that actually works.
The pattern is almost always identical. An account gets banned. The advertiser opens a new one, rebuilds the campaigns, relaunches. Two weeks later, sometimes three, the new account gets flagged too. They open another. Same result. By the fourth or fifth account, they’re convinced Meta has it out for them personally, that their product is impossible to advertise, or that the whole platform is broken.
None of those things are true. What is true is that Meta’s system flagged their original account for a reason, that reason was never addressed, and every new account they opened was contaminated by the same signals before the first ad even ran.
This is the distinction that most advertisers in restricted categories never figure out on their own: a ban is not an event. It is the end state of an accumulated risk profile. When you open a new account without clearing that risk profile, you are not starting fresh. You are inheriting the flags that killed the last account, applied to a new account ID that Meta’s system connects back to the original within days.
Meta doesn’t ban accounts in isolation. It flags entities. Your account is one signal among many that define that entity.
The fix is not a new account. The fix is a complete entity reset, every asset and identifier that Meta’s system uses to connect the new account back to the banned one. This article documents exactly what that means and what needs to change.
How Meta actually connects accounts
Before explaining what to change, it helps to understand what Meta’s system is actually looking at. Most advertisers assume Meta tracks accounts by login credentials and payment methods. Those matter, but they’re only two signals in a much larger fingerprint that Meta builds around every advertising entity.
Meta’s system is pattern-matching across a combination of signals simultaneously: device fingerprints from the browser or app used to manage the account, IP address history, the Business Manager the account lives in, the Facebook Page associated with the campaigns, the pixel ID firing on the destination domain, the domain itself, the payment method, the profile that created and manages the account, and the creative assets used in the campaigns.
What this means in practice: You open a new ad account. You log into it from the same computer you managed the banned account from. You associate it with the same Business Manager. You connect it to the same Facebook Page. You install the same pixel on the same domain. You run the same creatives.
Meta’s system sees: same device, same BM, same page, same pixel, same domain, same creatives. It does not see a new advertiser. It sees the same flagged entity operating under a new account ID. The new account inherits the risk profile of the banned one within days, often hours.
This is why opening a new account on its own never works for advertisers in restricted categories who have been flagged. The account ID changes. Nothing else does. The system that tracks the entity, not just the account, connects the dots immediately.
The complete reset: everything that needs to change
A genuine fresh start means replacing every asset that Meta’s system uses as an identifier. Not most of them. All of them. Changing six out of eight and leaving two the same is enough for the system to make the connection. The reset has to be complete.
Here is every element that needs to change, and why each one specifically matters.
Domain
The domain your ads send traffic to is one of the first things Meta’s system checks when reviewing a new campaign. A banned domain carries its history permanently. Register a completely new domain with no connection to the previous one, different registrar if possible, different WHOIS contact information, hosted on a different IP. Even a domain that was never formally banned but was associated with a flagged account carries residual risk. Replace it entirely.
Facebook Page
The Page your ads run from is a core part of your advertising entity on Meta. A Page that ran ads on a banned account is a flagged Page, even if the Page itself was never restricted. Create a completely new Page, built from scratch with a fresh name, different profile photo, different cover image, different about text, and ideally managed from a profile that has no history with the banned account. Do not migrate content from the old Page.
Ad account
The ad account itself obviously needs to be new. But “new” here means sourced properly, not just created fresh from your existing Business Manager. A brand-new account created by a flagged BM or managed by a flagged profile inherits risk before its first campaign runs. In restricted categories, accounts sourced through established agency infrastructure with clean spend history start from a fundamentally different trust baseline. This is why account sourcing through a vetted provider matters, not just account creation.
Personal profile
Every ad account is managed by a profile. If the profile managing the new account was also managing the banned one, Meta’s system connects them through the profile activity history. The new account should be created and managed by a profile with no history with the banned account, ideally a profile with its own clean advertising history and established account age. Fresh profiles created specifically to manage a new account carry their own heightened scrutiny risk.
Business Manager (BM)
The Business Manager is the container that holds your ad accounts, Pages, pixels, and payment methods. A flagged BM contaminates everything inside it. Create a completely new Business Manager, associated with a new profile, with no assets migrated from the old BM. Do not add the new ad account to an existing BM that held the banned account, even if that BM appears unrestricted. The association alone creates a connection Meta’s system will eventually surface.
Pixel
The pixel ID is one of the clearest technical identifiers Meta uses to connect advertising activity to a specific entity. A pixel that fired on a banned domain, or that was installed in a flagged BM, carries that history in its event data. Create a new pixel from scratch in the new BM, install it only on the new domain, and ensure the implementation follows the custom tracking architecture: firing only on the white page, never on the destination page, with server-side events as the primary data layer.
Creatives
Creative assets that appeared in banned campaigns are logged in Meta’s system. Running the same image, the same video, or even the same copy structure on a new account is one of the fastest ways to trigger a review flag on what would otherwise be a clean account. Produce entirely new creatives, new images, new video, new copy, new angles. Do not reuse, crop, recolour, or slightly modify existing banned creatives. The perceptual hash matching Meta uses can identify reused visual assets even after basic modifications.
Website photos and image metadata
This is the element most advertisers never think about and the one that catches them after they’ve changed everything else. Images on your website carry metadata: EXIF data containing the device used to take or edit the photo, timestamps, GPS coordinates in some cases, and editing software information. If the images on your new domain are the same files as on the banned domain, the metadata fingerprint is identical, even if the domain, hosting, and pixel are all new. Replace every image on the site with genuinely new files. If reusing any photos, strip all EXIF metadata before uploading, using a tool like ExifTool or an online metadata remover.
The rule that makes the checklist work: Every item on this list needs to change. Not most of them. All of them. Leaving one unchanged is enough for Meta’s system to connect your new account to the banned entity. The connection does not need to be obvious or direct. A shared pixel ID, a shared profile, a shared image file — any single link is sufficient for the system to inherit the previous risk profile onto the new account.
The connection map: how each asset links to the others
Understanding why all eight need to change becomes clearer when you see how Meta’s system links them. These are not eight independent checks. They are eight nodes in a connection graph, and finding any one of them that matches a banned entity allows the system to traverse the graph and find the rest.
Linked to pixel (installed on domain), to ad account (used as destination), and to WHOIS registration data. A banned domain found in the pixel event data of a “new” account immediately flags the account.
Linked to domain (where it fires), to BM (where it lives), and to ad account (which uses it for conversion tracking). Old pixel ID in a new BM is a direct connection to the banned entity.
Contains the ad account, the pixel, and the Page. If any of these assets inside the BM has a banned history, the BM itself is flagged. New assets placed inside a flagged BM inherit its status.
Connected to BM (admin role), to ad account (payment and management), and to Page (admin). A profile with banned account history that becomes admin of a new BM brings that history with it.
Linked to the ad account (campaigns run from the Page), to the profile (Page admin), and to the BM (Page asset). Reusing a Page that ran banned campaigns is one of the most reliable ways to transfer the ban.
Identified by perceptual hash independent of filename, format, or minor visual modification. The same image, even recoloured or cropped, produces a near-identical hash that Meta’s system matches against its library of flagged creatives.
EXIF data embedded in image files can identify the same device or editing workflow across a new domain’s assets. Identical metadata fingerprints across two domains are a signal that the same operator runs both, connecting a clean new domain to a banned old one.
The device and IP layer: what the checklist doesn’t cover
The eight assets above are the ones you control and replace directly. There are two additional signal layers that exist outside those assets and that need to be addressed separately: the device you use to manage the account, and the IP address you connect from.
Meta’s system builds a device fingerprint from the browser environment used to access Ads Manager: browser type and version, screen resolution, installed fonts, timezone, and a range of other signals that together produce a near-unique fingerprint. If you manage your new account from the same browser on the same device as the banned account, the device fingerprint matches, and the connection is made regardless of how clean every other asset is.
The cleanest solution is a completely separate device used exclusively for the new account, never used to access any asset associated with the banned account. If a separate physical device is not practical, a dedicated browser profile, completely separate from any profile used for the old account, with its own clean cookies, cache, and no shared sessions, is the minimum viable separation.
The IP address carries the same logic. Accounts managed from the same IP address as a banned account inherit a risk signal from that IP’s advertising history. A dedicated residential IP, different from the one used for the banned account, is the correct setup. VPNs are not a substitute: VPN IP ranges are well-known to Meta’s system and create their own flag signals. A clean residential IP, not shared with the banned account’s history, is what matters.
On payment methods: Payment card details are another connection point that most advertisers overlook. A credit or debit card used on the banned account should not be used on the new one. Meta connects payment methods to advertising entities and a card that appears on both a banned account and a new account is a direct link. Use a new card, from a new billing address where possible, for the new account’s payment method.
Why this happens repeatedly: the symptom fix cycle
The reason most advertisers go through four, five, or six bans before finding a solution is that each ban is treated as a problem with the account, not a problem with the operation. The account gets banned. A new account is opened. The new account gets banned. Another new account. The cycle continues because the fix being applied (new account) does not address the cause (inherited risk profile from incomplete entity separation).
Appealing instead of rebuilding
The appeal process on Meta for restricted category bans has a very low success rate and a significant cost: every appeal flags the account for manual review scrutiny, which accelerates future enforcement actions. Time spent appealing is time not spent on the correct fix. Unless there is a clear, obvious policy misapplication, appeal only with strong grounds. Otherwise, rebuild correctly and move on.
Partial resets
Changing the ad account and the domain but keeping the same BM, pixel, and profile. Or changing everything except the creatives. Each partial reset delays the inevitable flag rather than preventing it. The system finds the one unchanged connection, traverses the graph, and identifies the entity within days. A partial reset buys time, not safety. Only a complete reset produces a genuinely clean start.
Launching at full spend immediately
A new account with no spend history that immediately launches at high daily budgets attracts algorithmic scrutiny regardless of how clean the creative is. Warm new accounts with low-risk campaigns at modest spend for the first two to three weeks, establish a clean compliance record, and then scale. An account with two weeks of clean spend history faces less scrutiny on a high-risk campaign than an account that launched at scale on day one.
Using the same creative angles that caused the original ban
Even with a complete entity reset, running the same creative approach that generated the original flags will produce the same outcome, eventually. The reset removes the historical connection. It does not remove the policy risk of the creative itself. The complete reset needs to be paired with a genuine creative rebuild using compliant angles, not just the same content on new assets.
The reset is the start, not the finish
A complete entity reset, all eight assets replaced, new device and IP, new payment method, creates a clean foundation. What you build on that foundation determines how long it stays clean.
The advertisers who complete a proper reset and then continue running the same creative approach, the same white page architecture, and the same single-account structure that got them banned in the first place are typically back in the same cycle within two to three months. The reset gives you a clean start. The infrastructure and creative discipline you build on top of it determines whether you stay clean at scale.
A proper setup after a full reset includes multi-layer traffic filtering so that Meta’s review systems cannot reach your destination page directly, custom tracking so that your pixel never reports your destination URL back to Meta’s servers, compliant creative angles that do not rely on outcome claims or prohibited visual patterns, a parallel account structure across multiple accounts so that a single flag does not stop the operation, and a gradual spend ramp on new accounts that builds a clean compliance record before scaling.
The ban told you something was wrong. The reset clears the record. What you build next determines whether the lesson actually landed.
Stuck in the ban cycle
and not sure what’s connecting them?
We audit your full entity footprint, identify every connection point between your banned account and your current setup, and rebuild the entire operation from a clean foundation.
Book a strategy call