Meta Is Removing Placement Exclusions: Replace Blocking with a Quality Audit

If Ads Manager stops letting you exclude placements, platforms, devices, or operating systems, do not try to rebuild every block with bidding rules. Move to a post-delivery control loop: measure business quality by placement, inspect how each creative actually renders, and define the anomalies that justify intervention before budget moves.

One fewer exclusion requires more operating discipline. Spot when automation is buying cheap volume and poor business value before that pattern becomes the default way an account scales.

Social Media Today reported on August 21 that Meta is removing the option to exclude placements. Jon Loomer documented the account notice a day earlier and says it also covers exclusions by platform, device, and operating system; the effective date is not yet public.[1][2] If that notice is already appearing in an account, the change is not cosmetic. It changes where the media-buying work happens.

Assuming that "Advantage+ knows more than we do" leaves the account unsupervised. Rebuilding the old control with twenty rules is no better because it destroys a clean read on delivery. Let the system distribute inventory, then require business evidence wherever it concentrates spend.

What disappears and what remains your job

The notice Loomer reviewed says the ad set would no longer support four exclusion types: individual placements, whole platforms, device type, and operating system.[2] That does not mean every delivery decision vanishes or that every creative-objective combination is acceptable.

Your job shifts toward three things that Meta cannot determine from your business context:

1. Define which conversion is worth getting. A submitted form, a started message, and a closed sale are not the same signal. Optimize toward the easiest event and the system will find cheap inventory that produces it. That choice is human, not placement-level.
2. Give the system assets that work on each surface. A creative that gets cropped, covers the CTA, or looks recycled can still buy impressions and waste budget. The production work behind vertical safe zones matters more, not less, once placement selection is automated.
3. Catch decline with a post-click metric. Cheap CPM does not clear a placement. If its contact rate, qualified-lead rate, or CRM conversion rate falls, there is a quality problem even when Meta reports efficient delivery.

The platform optimizes toward the event you supply. The business has to live with what happens next.

The Mintec framework: four layers of post-delivery control

Use this instead of saying, "Block X because it always performs badly." The word that matters is evidence. Do not condemn a placement after a noisy week; review it against a signal agreed before budget moves.

Layer 1 prevents the expensive mistake: asking Meta for "more leads" when sales needs viable opportunities. Layer 2 prevents the next one: treating automatic resizing as the same thing as a piece designed for the surface. Layers 3 and 4 move the conversation from preference to evidence.

The 30-minute audit to run before changing anything

Use a 14-to-30-day window. If the account does not produce enough volume to assess a placement separately, do not force a conclusion. Keep it as a hypothesis and review the campaign aggregate.

1. Separate cost from quality

In Ads Manager, break out spend, impressions, CTR, cost per result, and results by placement while that breakdown remains exposed in the account. Then join the export to one downstream signal: contacted lead, attended demo, sales-accepted opportunity, net purchase, or margin. You do not need ten metrics. You need one the business accepts as valuable.

A simple example: Reels may carry a lower CPM than Feed, but that saving does not exist if Reels leads are much less likely to be contacted. Our August Meta CPM analysis explains why a platform-wide CPM can never become an account diagnosis. Cost is a clue. Quality decides.

2. Review a visual sample, not only columns

Open delivered ads in the surfaces taking the spend. Check five things: the product makes sense in the first second, critical copy is not under the interface, the CTA remains readable, the destination matches the ad, and the format does not make a Feed asset look like an accidental vertical crop.

That is not a brand-policing exercise for the creative team. It is conversion control. In automated delivery, a poorly adapted asset can still receive impressions because inventory exists for it. Do not conclude that the surface fails until you separate a placement problem from an asset problem.

3. Flag outliers with a written rule

Write the rule before opening Ads Manager. For example: "Review when a placement holds more than 25% of spend and delivers a qualified-lead rate 30% below campaign average in two consecutive cuts." Adjust the thresholds to the account's volume, but do not change the rule after seeing the data.

This guards against two bad habits: shutting down a placement because CPM looks high for three days, and letting a weak lead source scale because CPL looks pretty. It also lets you explain a decision to a client without hiding behind "the algorithm wanted it."

4. Keep an exception log

Every intervention needs one line: date, campaign, evidence, affected asset or signal, action, and next review date. If an exception improves sales acceptance but raises CPM, the result needs to stay visible. Otherwise, someone will re-enable the same structure in a month because they remember the cost but not the quality.

The log also helps you read measurement changes. Meta has already changed how it assigns conversions between click-through and engage-through. If you mix an attribution change with a delivery decision, you can give credit for an imaginary improvement to the wrong placement. Read the click-through versus engage-through change first and keep the quality criterion outside those attribution columns.

Value rules can test a hypothesis, not hide the problem

Loomer notes that value rules can raise or lower bids by placement, device, or operating system, but they are not the same as an exclusion.[2] That difference matters.

A value rule can make sense when evidence shows that a type of inventory produces a different value and the account has the volume to measure the change. It is not a solid answer when you use it to compensate for weak creative, a slow landing page, or a form that attracts curiosity clicks. In those cases, lowering the bid hides the cause and makes learning harder.

Our view: before applying a value rule, you should be able to write one sentence that supports it. "Attended demos from this delivery segment outperform the average in two comparable cuts" is measurable. "It has always gone badly there" is not.

When to escalate instead of optimize

Some limits should not become bidding experiments. If a surface creates a brand-safety risk, an offer incompatibility, a privacy requirement, or an approval problem, stop and review Meta's current policies and the account configuration. Do not represent a value rule as a safeguard that guarantees the ad will not appear there.

Escalate too when you cannot connect Meta delivery to a post-click quality signal. Without CRM data, server-side conversion capture, or a reliable quality marker, automation is free to optimize the easiest metric. Fix that signal before debating placements.

The media buyer is not disappearing. The job is moving up a level.

Meta is making a direction explicit that it has pushed for years: fewer inventory controls in the ad set and more automated delivery.[1][2] That does not make the operator irrelevant. It makes the habit of confusing an exclusion with a strategy irrelevant.

The remaining work is harder and more valuable: agree what quality means, make sure creative survives each format, connect the dashboard to business outcomes, and keep an exception record that can be defended. If preventive control shrinks, your audit has to improve. That is the operating system that protects the budget.

Sources

[1] https://www.socialmediatoday.com/news/meta-removes-option-to-exclude-ad-placements/828461 — Meta removes option to exclude ad placements [2] https://www.jonloomer.com/meta-removing-placement-controls-ad-sets — Meta is removing placement controls from ad sets