Should You Follow Meta's Opportunity Score? (2026)
Should I apply Meta's opportunity score recommendations?
Apply the ones that fix errors or match your goals, and dismiss the rest. Meta's opportunity score runs from 0 to 100 and reflects how many of its recommendations you have applied. Meta says even a high score does not reflect actual or future performance, and dismissing a recommendation leaves your score where it was.
Treat opportunity score as a to-do list written by Meta, not a grade on your results. Errors and warnings on it come first, because they block points until every one is cleared. The other suggestions are hypotheses: some fit your account, some cut against a margin target, a cash limit or a test you are running on purpose. Dismiss those without worry; Meta expects accounts with business constraints to stay below 100.
What does Meta’s opportunity score actually measure?
Meta describes it as a 0 to 100 scale of “how optimized your campaigns, ad sets and ads are”, based on how many Ads Manager recommendations you apply (About opportunity score). Three details matter more than the number itself:
- Recommendations are weighted. Each one carries a point value that Meta estimates from your business type, campaign objective, budget allocation and the lift it saw for that recommendation in its own experiments. In Meta’s example, fixing audience fragmentation is worth 35 points and Advantage+ placements 5.
- Errors count. Recommendations include errors and warnings, such as incorrect formatting, missing information or policy violations. If a campaign has several errors, you get no points until all of them are resolved.
- It is not a performance metric. Meta states that the score, including a high score, “does not reflect your actual or future performance”, that applied recommendations can help over time but this isn’t guaranteed, and that you should not turn off a campaign because its score is low.
The score appears in the Account Overview and Campaigns tabs of Ads Manager and when you create an ad. It does not show at all while the account has an account error. Meta also says the score is in development and may evolve, so the recommendation set changes over time.
Which recommendations does Meta make, and which should I apply?
Meta’s help pages describe several recommendation types that feed the score. The second column is what Meta says each one covers; the last two columns are how to judge it in a store account.
| Recommendation | What it covers (Meta) | Usually apply when | Dismiss or wait when |
|---|---|---|---|
| Errors and warnings | Issues such as incorrect formatting, missing information or policy violations (About opportunity score) | Always review first; errors block points | Never ignore an error; fix or remove the item |
| Combine ad sets | Similar ad sets that fragment one audience (Combine ad sets) | Several similar ad sets split one audience | The ad sets are a deliberate test with separate audiences |
| Auction overlap | Ad sets whose ads enter the same auctions (Understand auction overlap) | The overlapping ad sets are learning limited | The overlap is part of a planned test |
| Audience expansion | A larger suggested audience for up to 3 ad sets (Audience expansion) | A small audience is pushing cost per result up | Your audience is narrow on purpose, such as a retargeting list |
| Creative fatigue | A new ad, a wider audience or Advantage+ creative (Creative fatigue) | Cost per result is rising as the same people see the ad again | The ad set is new and still collecting results |
| Budget | Budget changes for eligible campaigns (Campaign and ad set budgets) | Cash and margin support more spend | The raise is large enough to restart learning, or cash is tight |
When should I apply a recommendation?
Apply it when it repairs something broken or closes a gap you can see in your own data. Three cases come up most in store accounts.
- Errors and policy warnings. These are problems with the ad or campaign, not suggestions. Fix them first, because the account earns no points for a campaign until all of its errors are cleared.
- Audience fragmentation and auction overlap. When ad sets share an audience, Meta enters only the ad with the highest total value in each auction, so the others can underspend or stay in learning (Understand auction overlap). If the ad sets are not a planned test, combining them is usually right. See how do I check and fix audience overlap on Facebook?
- Audience expansion. Meta says a small audience can push cost per result to at least twice what your past ad sets cost, and it suggests a larger audience for up to 3 ad sets (About audience expansion recommendations). The suggestion leaves your location, custom audiences and exclusions alone.
Creative fatigue recommendations sit close behind. Meta shows a “creative fatigue” status when an ad’s cost per result is at least twice what your past ads cost, and “creative limited” when it is higher but under twice (Creative fatigue recommendations). That is a claim you can check against your own frequency and cost columns before you act; the creative fatigue guide covers how.
When should I dismiss a recommendation?
Dismiss it when it conflicts with a constraint Meta cannot see. Meta allows for this directly: if you can’t apply some recommendations “due to business constraints”, your score will stay below 100 (About opportunity score).
- A budget increase you cannot fund. “Scale your high-performing campaign” knows the campaign’s results, not your cash position or contribution margin. A large jump can also send ad sets back into learning: Meta’s example is a move from $100 to $1,000 (About significant edits). Read which edits reset the learning phase on Facebook ads? before accepting one.
- A goal change that fights your measurement. If you chose value optimization because order values vary widely, a prompt to maximize the number of conversions moves the campaign away from what you judge it on. The trade-off is covered in should I use value optimisation or purchase optimisation?
- A merge that would end a test. Two ad sets with overlapping audiences may be an A/B test you set up deliberately. Finish the test before combining them.
Dismissing works per campaign. You may dismiss a recommendation on one campaign and see the same one on another, so review each campaign rather than assuming a dismissal covers the account.
How do I review opportunity score recommendations in Ads Manager?
Meta shows the score in the Account Overview and Campaigns tabs of Ads Manager, and when you create an ad (About opportunity score). Work through it in this order:
- Open Ads Manager > Account Overview and find your opportunity score and its list of recommendations.
- Read the point value next to each recommendation, and note which campaign or ad set it applies to.
- Sort the list: errors and warnings first, then audience and creative items, then budget and goal changes.
- Check each budget or goal change against your margin, cash and any running test before you touch it.
- Apply the ones that pass, following what Meta shows you to do on each recommendation.
- Dismiss the rest; each moves to the Dismissed tab in Account Overview and your score stays the same.
- Open Ads Manager > Campaigns, check the score there, and repeat for each campaign, since a dismissal covers one campaign only.
After you apply a change, judge it on your own numbers. Compare cost per purchase, purchase value and store revenue for equal date ranges before and after, not the score.
Does a low opportunity score hurt my ads?
Meta’s guidance is that a low score means there may be more recommendations you could apply, and that you should not switch off a campaign because of it: the score says how well-optimized a campaign is, not whether it is performing poorly (About opportunity score). Meta also says applied recommendations can help over time, but that this isn’t guaranteed.
The practical risk runs the other way. Chasing 100 means accepting every budget increase, goal switch and merge the system proposes, including ones built on a goal you do not share. For the full free audit Meta’s own tools allow, see what are the best Facebook ads audit tools, free and paid? and the Meta ads audit checklist.
Can software help?
Bach.ai (by Wittelsbach) connects to a Meta ad account, audits it daily, finds revenue leaks with an estimated revenue impact and proposes fixes, which works as a second opinion beside Meta’s list. On every plan it applies a change on Meta only after you approve it.
FAQ
Does a high opportunity score mean my ads are performing well?
No. Meta says the score, including a high score, does not reflect your actual or future performance. A high score means you have applied most recommendations and cleared errors and warnings, or that Meta has nothing to suggest right now. Judge performance on cost per purchase, purchase value and your store’s revenue instead.
Will dismissing a recommendation lower my opportunity score?
No. Meta says your score will not change when you dismiss a recommendation, and the recommendation moves to the Dismissed tab in Account Overview. The trade-off is that a dismissed recommendation’s points stay unearned, so an account with real business constraints will sit below 100 for good.
Why did my score not change after I applied a recommendation?
Two common reasons from Meta’s help page. If a campaign has several errors, it earns no points until all of them are fixed. And a recommendation that creates something new, such as a campaign with suggested presets, only adds its points once that campaign is active.
Why can’t I see my opportunity score?
Meta says the score does not show while your ad account has an account error, so resolve that error first. On desktop the score sits in Account Overview, above the Campaigns table and in ad creation; the Ads Manager app has its own route to the same recommendations.
Sources
Sources: Meta Business Help Center, About opportunity score, Understand auction overlap, Combine ad sets to reduce audience fragmentation, About audience expansion recommendations, Creative fatigue recommendations, About campaign and ad set budgets and Significant edits and learning phase (checked 2 Oct 2026).