TL;DR
- Google defines optimization score as an estimate, from 0% to 100%, of how well a Google Ads account is set to perform. It is shown at campaign, account and manager account level.
- Google calculates it in real time from account and campaign statistics, settings and status, the impact of available recommendations, and recent recommendations history.
- Google states that optimization score is not used by Quality Score, and it is not among the six Ad Rank factors Google lists.
- Applying or dismissing a recommendation both change the score. A dismissed recommendation can return after a preset period if the campaign is still eligible.
- Google's Partner requirements set a minimum optimization score of 70% for the agency's registered Ads manager account, checked daily.
Google Ads optimization score is Google's estimate, shown as a percentage from 0% to 100%, of how well an account is set to perform. It is recalculated whenever a recommendation is applied or dismissed, and Google states that it is not used by Quality Score, so the number tracks outstanding recommendations. The score uplift on each one reflects, when possible, Google's estimate of its impact on overall campaign metrics, which is not the same as a measured business result.
That distinction decides how the score should be used. A 100% score means no recommendation is left open. It does not mean the campaigns are profitable. This guide explains how the score is calculated, what it does and does not influence, how to handle recommendations one at a time, which automatic changes deserve a second look, and where the score fits into the Google Partners programme.
What is optimization score in Google Ads
Google's help documentation defines optimization score as an estimate of how well a Google Ads account is set to perform. Scores run from 0% to 100%, and Google describes 100% as meaning the account can perform at its full potential.
The score is available at three levels: campaign, account and manager account. Google shows it only for active Search, Display, Video Action, App, Performance Max, Demand Gen and Shopping campaigns. A paused campaign does not get one. Advertisers who run several of these types together, for example Search alongside Google Shopping campaigns, will see a score for each campaign and one for the account as a whole.
The score sits at the top of the Recommendations page, above a set of recommendation cards. Each card carries a percentage called the score uplift, which Google describes as the possible score improvement from following that recommendation. Google's own example is a recommendation type with a 7% uplift: applying all of it raises the account's score by 7%.
How optimization score is calculated
Google says the score is calculated in real time from the following inputs:
- the statistics of the account and its campaigns
- their settings
- the status of the account and campaigns
- the relevant impact of the recommendations currently available
- recent recommendations history
Two things follow from that list. Because the calculation is in real time, the score can move when nobody has touched the account. New performance data or a changed setting can produce a new recommendation, and an open recommendation lowers the score until someone deals with it.
Because the score reflects the relevant impact of each recommendation, the cards are not worth the same amount. A recommendation that Google estimates would have a larger effect carries a larger uplift, so one bidding suggestion can be worth more than several small asset suggestions put together.
Google does not publish the formula behind the uplift figures. What is documented is the list of inputs above. Anything more specific, such as a fixed weight per recommendation type, is guesswork. The recommendations themselves are generated from the account's performance history, its campaign settings, and Google search volume and trends, so the list is specific to each account.
Does optimization score affect Quality Score or Ad Rank
No. Google's help page states that optimization score is not used by your Quality Score. Google describes Quality Score as a diagnostic tool for comparing ad quality with other advertisers, and says optimization score provides a more holistic view of the account's health.
Ad Rank is a separate calculation again. Google's documentation lists its factors: the bid, the quality of the ads and landing page, the Ad Rank thresholds, the competitiveness of the auction, the context of the person's search, and the expected impact of ad assets and other ad formats. Optimization score is not on that list. A higher score does not, by itself, lower a cost per click or move an ad up the page.
The link between the score and performance is indirect. Applying a recommendation changes something real in the account: a keyword is added, a bid strategy is switched, an asset is attached. That change enters the auction and can move results in either direction. The score only records that the recommendation is no longer open. Dismissing a recommendation also changes the score, and it changes nothing in the campaigns, which shows how loosely the two are tied.
Should you apply every recommendation
No. A recommendation is a suggestion generated from account data. It has no view of profit margin, stock levels, how many leads the sales team can handle, or the monthly ceiling agreed with finance. The useful question for each card is what applying it would change, and whether that change suits the business.
| Recommendation | What applying it changes | Question to ask first |
|---|---|---|
| Raise your budgets | The campaign's budget goes up, so it can spend more | Is the extra spend affordable at the current cost per conversion? |
| Add broad match keywords | Ads become eligible for a wider range of searches | Is conversion tracking reliable, and is someone reviewing search terms? |
| Set a target CPA | Bidding works toward a cost-per-conversion target | Is the suggested target one the business can live with? |
| Remove redundant keywords | Keywords Google flags as redundant within the same ad group, destination and bidding strategy are removed | Is any of them kept separate on purpose, for reporting? |
Budget recommendations carry the plainest trade-off. Google describes "Raise your budgets" as a fix for a limited budget so that ads keep running on the busiest days. That may suit a profitable campaign held back by its budget. For a campaign that is not yet paying for itself, the same change buys more of the same result. The score rises either way, so the score cannot be the reason to accept.
Broad match expansion widens the set of searches that can trigger an ad. Google describes the "Add broad match keywords" recommendation as broad match versions of existing keywords intended to help grow conversions. Whether that works depends on how well the account can tell a good click from a poor one. With accurate conversion tracking and a habit of reading the search terms report, wider matching can be tested safely. Without them, the account is paying to find out.
Bidding recommendations such as "Set a target CPA" hand more control to automated bidding. The trade-off is between convenience and oversight: the target has to be one the business accepts, and the campaign needs conversion data the bidding can rely on.
Dismissing is a legitimate answer. If a recommendation does not fit, declining it is the correct decision, and it clears the card from the score in the same way.
How to dismiss recommendations
- Open the Recommendations page from the Campaigns menu.
- Find the card and select "View recommendation" to read the detail.
- If it is not relevant, select "Dismiss all".
Scope matters. Dismissing a recommendation at campaign level removes that type for the chosen campaign only, and other campaigns can still show it. Dismissing at account level stops that type from appearing in any campaign in the account for a preset period.
A dismissal is not permanent. Google says recommendations can be dismissed for 30 days or 60 days depending on the reason given, and that a dismissed recommendation can resurface after a preset period if the campaign is still eligible. Seeing a card return a month or two later is normal.
To undo a dismissal, open the Recommendations page, select the "Dismissed" filter at the top, open the three-dot icon on the card and choose "Undismiss all". Keep a short written note of why each recommendation was declined. When the card comes back, or when someone new takes over the account, the reasoning is already there.
Auto-applied recommendations: which to turn off
Auto-apply lets Google Ads apply selected recommendation types on a recurring basis, without anyone approving each change. The setting is reached from the top bar of the Recommendations section. There you can choose specific recommendations or one of two bundles, which Google names "Maintain your ads" and "Grow your business". Only the types selected in those settings are applied automatically. Auto-apply works at account level only, not per campaign, and Google says it can be turned on or off at any time.
Which types to switch off is a question of where human review is worth the time. One way to sort them:
- Types that change which searches an ad can appear for. Adding keywords or widening match types alters who sees the ads. If these run automatically, new keywords go live before anyone has read them, so manual review is the safer default.
- Types that change how bids are set. A bidding change is a commercial decision about what a conversion is worth. Leaving it on auto-apply means that decision is made without the person who knows the margins.
- Types that tidy the account. Housekeeping changes such as removing redundant keywords look minor, but Google notes that this recommendation can include keywords across different match types. Review what it would remove before switching it on, and make sure someone still reads the log.
That log is the History tab on the Recommendations page, which shows a summary of recently applied recommendations. For each individual change, Google points to the Change History page. A simple routine is to open History on a fixed day each week. Entries nobody on the team recognises are the sign that auto-apply is on for something it should not be.
Optimization score and Google Partner requirements
The score matters to agencies for a reason unrelated to any single campaign: it is part of the Google Partners programme. Google's help page on becoming a Partner states that the Ads manager account registered with Google Partners needs a minimum optimization score of 70%. Performance is one of three requirement areas, alongside spend and certifications, and Google says the requirements are checked daily. Programme rules change, so the current figure should be confirmed on Google's Partners help page.
Two points follow for an advertiser working with an agency. The threshold applies to the agency's manager account, not to each client account, and it is well short of 100%. Because dismissing a recommendation also changes the score, an agency can meet the requirement while declining suggestions that do not suit a client.
The badge shows that an agency meets Google's programme requirements. It does not describe the results in any particular account. If a change is proposed "for the score", it is reasonable to ask what the recommendation alters in the account and why it suits the campaign.
What this means in the Thai market
For advertisers in Thailand the mechanics are identical, and a few local details affect how recommendations should be read. Accounts billed in baht see budget suggestions in baht, and the figure should be compared with the monthly budget the business has approved before anything is accepted.
Keyword recommendations need a closer read in Thai. Thai is written without spaces between words. After a broad match recommendation is applied to Thai keywords, the search terms report for the following weeks shows what the ads were matched to, and that report is the evidence for keeping or reversing the change.
Suggested ad text and assets should be read by a Thai speaker before they are applied, to check that the wording is natural and the claims are accurate for the business. The same care applies to visual formats, including display advertising, where an asset added in a hurry represents the brand on other websites.
Where an agency or a freelancer runs the account, three questions are worth asking whoever manages it: is auto-apply switched on, for which recommendation types, and who reads the History tab.
Frequently asked questions about Google Ads optimization score
What is a good optimization score in Google Ads?
There is no score that guarantees good results, because the score measures open recommendations and not profit. Google's Partner requirements set a 70% minimum for the registered manager account. For an advertiser, a lower score with every recommendation reviewed is a sounder position than 100% reached by accepting everything.
Does dismissing a recommendation hurt the account?
No, dismissing changes the recommendation list and the score, and nothing in the campaigns. Google states that applying or dismissing recommendations changes the overall optimization score. A dismissed recommendation can reappear later if the campaign is still eligible for it.
Why did the score drop when nothing was changed?
The score is calculated in real time, so it moves when the account data moves. New statistics or a change in campaign status can generate a new recommendation, and an open recommendation lowers the score until it is applied or dismissed.
Can auto-applied recommendations be turned off?
Yes, Google says auto-apply can be turned on or off at any time in the auto-apply settings, which are reached from the top bar of the Recommendations section. The History tab shows what was applied recently, and the Change History page shows each individual change.
Is optimization score the same as Quality Score?
No, they are separate measures. Google states that optimization score is not used by Quality Score. Quality Score is a diagnostic of ad quality, and optimization score is an account-level estimate tied to recommendations.
How to use the score
The score is helpful as a checklist, because it surfaces settings and gaps that are easy to miss. As a target, it rewards accepting changes without asking what they cost. For help deciding which recommendations suit an account, see the Google Ads service page, or read how campaigns are run on the Google Ads management page.







