Google Ads automation is Smart Bidding, broad match and the recommendations Google can apply to your account on your behalf. It works well with a person piloting it. When nobody reviews the changes, auto-applied recommendations can quietly redirect where your existing budget is spent, and keep resetting what the system has learned.
The upside is real. According to Google Ads Help, broad match with Smart Bidding can deliver "~25% more conversions (in Target CPA campaigns)", a figure Google attributes to its own internal data from 2020. In the same breath, Google notes that "additional budget may be required" to reach those estimates.
This guide is for business owners and in-house marketers who see a steady stream of recommendations in their account and want the benefits of automation without handing over the budget. It is part of our series on unsupervised AI in digital marketing.
Contents
- What is Google Ads automation?
- What Google Ads automation does well
- How auto-applied Google Ads recommendations work
- Where your budget goes when nobody reviews the changes
- What your optimization score does and does not tell you
- Optimizing PPC campaigns vs Google Ads automation: you need both
- How to use Google Ads automation without losing control
- How Ascend Group Media pilots Google Ads automation
- Frequently Asked Questions
- Resources
What is Google Ads automation?
Google Ads AI automation works in three layers. Each one makes decisions you used to make by hand.
- Smart Bidding: Google sets your bids for each auction, aiming at a goal such as target CPA, target ROAS or Maximize conversions.
- Broad match: Google matches your keywords to a wider range of searches it judges to be relevant.
- Recommendations: Google suggests changes to your account. If you switch on "Automatically apply recommendations", it applies eligible ones for you.
Automated rules are a separate category: you set them up yourself, for example to pause a campaign or adjust budgets on a schedule, and they follow your instructions. The three layers above work from Google's judgment, so they get the most value from a regular review.
What Google Ads automation does well
Automation handles more volume than any person can. It prices bids auction by auction and finds searches you never thought to add as keywords. Google's own figures make the case: broad match paired with Smart Bidding delivered "~25% more conversions (in Target CPA campaigns)" and "~12% more conversion value (in Target ROAS campaigns)", based on what Google Ads Help cites as "Google internal data, 2020".
It is also built to keep working while it adjusts. During Smart Bidding's learning period, Google says "your ads continue to run and serve as normal" (Google Ads Help).
And auto-apply comes with a guard many advertisers don't know about. "Auto-applying recommendations won't increase your budget," says Google Ads Help, adding that "Budget raising isn't included in this iteration of auto-apply recommendations."
So the risk is not a bigger bill. It is where the bill goes.
How auto-applied Google Ads recommendations work
Google Ads recommendations sit in your account whether you act on them or not. Auto-apply changes who acts. According to Google Ads Help, "When you turn on 'Automatically apply recommendations', the recommendations will apply regularly," and "we'll know which recommendations are eligible for you on a daily basis."
That is the part worth pausing on. Once a recommendation type is switched on, changes can land in your account without anyone on your side approving each one. Many are sensible housekeeping. Some move money.
Which recommendations can apply themselves
Google's list of auto-apply types is long. These are the ones that change where spend lands:
| Recommendation | What it changes | Review before switching on |
|---|---|---|
| "Add broad match keywords" | Matches your ads to a wider set of searches | Yes: confirm the campaign uses Smart Bidding, then read search terms |
| "Add keywords" and "Add keywords (Smart bidding)" | Adds new searches to bid on | Yes: check each keyword fits what you sell |
| Expand reach with Google search partners | Shows search ads beyond Google Search | Yes: compare partner results with Google Search results |
| Display expansion | Extends search campaigns onto Display placements | Yes: check lead quality from the new placements |
| "Add dynamic search ads" | Lets Google match searches to pages on your website | Yes: check which pages and searches it targets |
| Adjust CPA/ROAS targets | Changes the cost or return your bidding aims for | Yes: check the new target against your margins |
| Switch bid strategies (for example Maximize conversions, Target ROAS) | Changes how every bid is set | Yes: agree the goal first and allow time to settle |
| Remove "redundant, non-serving, or conflicting" negative keywords | Removes exclusions you added | Yes: confirm each exclusion is truly no longer needed |
Recommendation types from Google Ads Help, "About applying recommendations automatically".
One recent change is good news for control: as of January 26, 2026, "Add responsive search ads" no longer auto-suggests or applies new ads, and it has been removed from the Auto Apply settings page, according to Google Ads Help.
Where your budget goes when nobody reviews the changes
Think of your daily budget as water running through a set of pipes. Auto-apply can't add water. It can add pipes, widen them and change the pressure. Each of those changes has a sensible purpose, and each one deserves a person checking that the water still reaches the right place.
Broad match and expansion change where existing spend lands
Because auto-apply won't raise your budget, every expansion recommendation draws on money you already set aside. New broad match keywords, search partners, Display expansion and new targets all redistribute that same budget across new searches, sites and placements. When the new traffic converts, that is automation earning its keep. When it doesn't, spend that once funded your strongest keywords is now funding tests nobody chose.
Google draws a clear line on broad match: "we strongly recommend only using these with campaigns using Smart Bidding strategies" (Google Ads Help). In our experience, the first thing to check after any broad match change is that the campaign is on Smart Bidding, because that is the pairing Google designed it for.
Bid strategy changes can add volatility
Bid strategy recommendations deserve the most care. Matthew Umbro, writing in Practical Ecommerce in May 2021, put it plainly: "Bid strategy recommendations cause too much volatility when automatically applied." He also noted that a CPA-increase recommendation "could also increase spending beyond the advertiser's budget."
Read alongside Google's budget guard, the lesson is clear. Auto-apply leaves your budget setting alone, but a higher CPA target changes how much the system is willing to pay for each conversion. That is a decision about your margins, and it belongs with someone who knows them.
Frequent changes keep resetting Smart Bidding learning
Smart Bidding learns best from a stable goal. Google explains that "Learning status will show when you make major structural changes," and that "It never lasts more than 7 days." Beyond that window, Google says the system takes one to two conversion cycles to converge on a new goal. Budget and target changes don't trigger Learning status, according to Google Ads Help.
One planned structural change is easy to absorb. A run of unreviewed structural changes is different: the system keeps recalibrating to new conditions instead of improving against a settled one. For a business with a long sales cycle, each round of settling takes longer still.
Less visibility means more to check
Automation also means fewer decisions you can see by default. Trade press has followed this closely: MediaPost reported in February 2025 that "Performance Max's 'black box' has been the most frustrating for many advertisers due to its lack of available data" (MediaPost). Google has since added more reporting, and we cover how to put it to work in our guide to Google Performance Max. The principle carries across every campaign type: the less you see by default, the more deliberately you look.
What your optimization score does and does not tell you
Your optimization score is easy to treat as a grade. Google describes it more modestly: "Optimization score is an estimate of how well your Google Ads account is set to perform." Google also states that "Optimization score is not used by your Quality Score" (Google Ads Help).
How the score moves is telling. Each recommendation shows its projected score uplift, and both applying and dismissing recommendations change the score. Dismissing can raise it to 100%. So a perfect score can mean you applied everything, or that you reviewed everything and declined what didn't fit your business. For account optimization, the score is a useful to-do list. The scoreboard is your cost per lead, cost per sale or revenue.
Should you follow your Google Ads rep's recommendations?
Google Ads rep recommendations come from people who know the platform well and can spot features you haven't tried. Treat them the way you would treat any recommendation: valuable input, checked against your goals and your numbers before anything changes.
A good question for every call is how a suggestion will move the number every PPC campaign should answer to, whether that is cost per lead or return on ad spend. If it should improve that number, test it. If the main benefit is a higher optimization score, take a closer look before saying yes.
Optimizing PPC campaigns vs Google Ads automation: you need both
This is not a choice between a person and a machine. Automation handles bids and matching across more auctions than anyone could price by hand. A person does what the system can't: sets the business goal, decides what counts as a conversion, reviews changes before they touch the budget and chooses what the account should learn from.
Think of an aircraft. The autopilot flies a long, steady route more precisely than any pilot. The pilot chooses the destination, watches the instruments and takes the controls when conditions change. That partnership sits at the heart of what PPC management involves today.
If you are still weighing whether the channel suits your business at all, our take on whether Google Ads are worth it for small business is a good place to start.
How to use Google Ads automation without losing control
Here is how to use Google Ads automation so it keeps finding conversions while you keep control of the budget. It is the routine we would run on any account:
- Review your auto-apply settings. Keep spend-relevant types off until someone has reviewed them. Google says the feature "can be turned on or off at any time by updating your auto-apply settings" (Google Ads Help).
- Check History and Change History on a set schedule. Applied changes appear in both (Google Ads Help). Disable anything you did not intend.
- Use broad match only with Smart Bidding. Plan for the extra budget Google says "may be required" to reach its estimates.
- Batch structural changes. Group them so Smart Bidding settles once instead of relearning every week.
- Read your search terms after any match type or expansion change. Add negative keywords where traffic doesn't fit what you sell.
- Judge every recommendation against your business number, not the optimization score.
How Ascend Group Media pilots Google Ads automation
At Ascend Group Media, we use Smart Bidding, broad match and recommendations wherever they help an account grow. What we add is the pilot: an experienced specialist reviews every change before it touches the budget, checks it against the client's revenue goals and times structural changes so the bidding has room to learn.
What we check first on any new account is the auto-apply settings, because they show which decisions are already being made automatically. That review is built into our paid advertising management. The same principle holds across Google and Meta ads automation, which we cover in our guide to Meta Advantage+.
Google Ads automation is at its best with a person steering it. If you would like a second pair of eyes on your account's auto-apply settings, get in touch with Ascend Group Media and we will walk through them with you.
Frequently Asked Questions
Should I turn on auto-apply recommendations in Google Ads?
You can, selectively. We suggest keeping spend-relevant types, such as broad match keywords, search partners, Display expansion, target adjustments and bid strategy switches, off until a person has reviewed them. Google says the feature "can be turned on or off at any time by updating your auto-apply settings", so you can start narrow and widen it as the account proves each change works for your business.
Can auto-applied recommendations increase my Google Ads budget?
No. Google states that "Auto-applying recommendations won't increase your budget" and that "Budget raising isn't included in this iteration of auto-apply recommendations." What auto-apply can change is where that budget goes. New keywords, broader matching, partner sites, Display placements and bidding targets all redistribute the spend you already have, which is why a regular review still pays off.
How do I turn off auto-applied recommendations?
Google says auto-apply "can be turned on or off at any time by updating your auto-apply settings." You can switch recommendation types off from the Manage tab or disable them from the History tab. Applied changes are listed in History and in your account's Change History, so you can see what has already been applied and decide whether to keep it.
Does optimization score affect Quality Score?
Google states that "Optimization score is not used by your Quality Score." Google describes optimization score as "an estimate of how well your Google Ads account is set to perform." Use it as a prompt to review recommendations, and judge each one against your own business results before applying it.
How long does the Smart Bidding learning period last?
Google says Learning status "never lasts more than 7 days" and shows "when you make major structural changes." Settling on a new goal can take longer: Google says the system takes one to two conversion cycles to converge. During learning, "your ads continue to run and serve as normal." Budget and target changes don't trigger Learning status, according to Google.
Is broad match safe to use?
Broad match works best under the conditions Google designed it for. Google says "we strongly recommend only using these with campaigns using Smart Bidding strategies," and notes that "additional budget may be required" to reach its performance estimates. Pair it with Smart Bidding, set a budget you are comfortable testing and read your search terms regularly, adding negative keywords where traffic doesn't match what you sell.
Should I follow my Google Ads rep's recommendations?
Treat them as useful input. Google Ads reps know the platform and often surface features worth testing. Before applying a suggestion, ask how it will affect your cost per lead, cost per sale or return on ad spend, and check it against your margins. If it improves the number your business runs on, test it. If the main benefit is a higher optimization score, review it more closely first.
Resources
- Google Ads Help: About applying recommendations automatically
- Google Ads Help: auto-apply settings and history (answer 10276359)
- Google Ads Help: About optimization score
- Google Ads Help: About Smart Bidding learning
- Google Ads Help: Grow your Smart Bidding campaigns with broad match
- Practical Ecommerce, Matthew Umbro, May 3, 2021: When to auto-apply Google Ads recommendations
- MediaPost, Laurie Sullivan, February 13, 2025: Google addresses ad platform questions on transparency
From the blog.
Let’s put this to work for you.
Tell us where you want to go. We’ll map the fastest path to revenue, no obligation.