Meta Manual Ad Placements: What Changes in 2026

Meta is removing manual ad placement selection at the ad set level for Sales, Leads, Engagement, and App promotion campaigns, according to Jon Loomer’s October 27 report on Meta’s updated guidance. Awareness and Traffic retain it. The change is rolling out gradually, so the controls you see may differ between accounts. For advertisers, the immediate job is to check creative across formats and judge placement results against business outcomes, rather than cost per result alone.

This article reflects reporting available on October 29, 2026. Check the settings in your account before changing a campaign.

Which campaign objectives lose manual placements?

The reported changes apply by campaign objective:

The performance goal does not determine this distinction. A Sales campaign optimizing for landing page views can lose manual selection while a Traffic campaign using that goal retains it. Manual platform selection and mobile-versus-desktop controls are also affected. These details are described in Jon Loomer’s October 27 update, which links to Meta’s announcement.

That does not mean switching every campaign to Traffic to keep a familiar setting. Start with the business outcome you need the campaign to optimize toward. Placement control is one part of that decision.

Which controls remain?

Three different controls address different problems. Keeping them separate will save you from applying a bidding adjustment when you actually need an exclusion.

Account controls are gaining objective-specific scoping, according to the same update. Availability and the list of supported exclusions need checking in your account. If an exclusion is required by a contract or brand-safety policy, confirm that the supported control actually prevents delivery. A lower bid does not satisfy that requirement.

A cheap lead can be an expensive result

Imagine two placements each spending $1,000. These figures are illustrative, not YieldBI customer results.

Placement A looks better in a report sorted by cost per lead. Placement B produces more than twice as many qualified leads for the same spend. The calculation is simple: divide spend by qualified leads, rather than by every form submission.

You can only make that comparison if qualification data reliably connects to the relevant placement. A campaign-level sales total does not prove that one placement produced better customers. If the connection is missing, fix the measurement before using the example as a reason to change bids.

Also allow time for leads to qualify. Comparing yesterday’s leads with last month’s completed sales mixes different stages of the buying process. Our guide to turning performance data into action explains why the decision rule and observation window belong together.

Prepare the creative before blaming the placement

A square product ad may look fine in Feed and lose a product detail in a vertical crop. Check the actual composition: can someone still see the product, understand the offer, and read the call to action?

Prepare square, portrait, and vertical versions where appropriate, checking each placement’s current specifications. Changing the canvas size alone may leave the important material in the wrong place. Keep essential copy away from areas covered by interface controls.

YieldBI’s static image ad workflow starts with a platform and placement before composition and generation. Use that order when preparing variations, then inspect each output before approving it. Approval keeps a creative; it does not publish an ad.

Better cropping fixes presentation. It cannot fix a form that attracts unsuitable leads or a conversion event that measures the wrong action.

Use value rules when the evidence supports them

If a placement consistently produces outcomes worth less to your business, an eligible value rule can express that difference through a bid adjustment. Choose the adjustment from your evidence rather than copying someone else’s percentage.

In YieldBI, Value Rules support criteria including placement and device platform. Review rule order and linked ad sets before saving. The first matching rule wins, so a broad rule above a specific one can change which adjustment applies. Updating a shared set can also affect several campaigns.

After a change, review delivery and business outcomes over a period that fits your conversion cycle. Avoid changing creative, bids, and the qualification definition together if you want to understand which change helped.

What about campaigns optimizing for an earlier action?

A booking can be canceled, and a trial can end without a paid subscription. When the event you optimize toward is a proxy for the outcome you want, its initial cost tells only part of the story.

YBO (YieldBI Bid Optimization) uses available campaign and business-outcome data to suggest adjusted bids for proxy-event conversions. It does not restore manual placements or imply placement-specific bid control.

Keep monitoring focused on decisions

YieldBI’s Ad Rules guide describes notifications and supported pause or resume actions. Start with an alert if a matching condition needs investigation. Saving a rule and activating it are separate steps, and options marked Soon are not available actions.

Ad Rules respond to conditions; Value Rules adjust eligible bids. Neither restores manual placement selection. If performance worsens, use a diagnosis such as what to optimize in Meta ads before assuming placement is the cause.

For your next review, pick one affected campaign. Check its available controls, inspect the creative crops, and compare initial conversions with the downstream results you can actually attribute. Change the setting that addresses the evidenced problem, then record what happened. That is a more useful response than rebuilding an old exclusion list from memory.

Sources

• Meta Business Help Center: placement settings. Meta’s guidance; access may require a login.
• YieldBI product guidance linked above, updated October 21, 2026. Documentation supports the described workflows; account-specific availability still needs checking.