Auto-applied recommendations in Google Ads: let them run or review every one?

Plenty of accounts have auto-apply on without anyone ever deciding that. Then a budget bump or a broad match keyword shows up that nobody asked for. Google calls these changes low risk, but low risk for whom?

If the account already converts, I'd turn off anything that touches bids, budgets or match types and only let small asset clean-ups through. A new account with thin data is different, since the automation can sometimes find useful things faster than a person working through the list every Monday. Microsoft Ads has a similar auto-apply setting, and the same logic applies there.

Which type of recommendation would you let through without looking?

Asset clean-ups are the safe bucket, but ad strength is still worth watching. Google's "improve ad strength" and extension suggestions can quietly add pinned headers or sitelinks that pull impressions away from the copy you wrote. The account looks healthier on paper while the messaging drifts. For a new account, I'd let the extension suggestions through and leave pinning and responsive ad rewrites for a human to read first.

On ad strength, the rating only counts how many unique assets you have and how varied they are. It doesn't check whether any of it is true. For an online shop that matters. Say a store offers free shipping over a certain amount, and an auto-added headline says "Free Shipping" on every ad. The rating goes up, but the ad now promises something checkout doesn't deliver. That's where I'd look first. Check any new headline or description against your feed price and your shipping and returns policy before you accept it. And if you want the copy to stay in your hands, turn off auto-created assets at the campaign level.

Agreed, and the feed check is the step most people skip. I'd pair it with a measurement step. Auto-applied changes show up in the change history with a date, so when a new headline goes live, compare CTR and purchase rate for the two weeks on either side. If CTR rises but purchases fall, the headline is overpromising, and the ad strength score won't catch that.

Agreed on the before-and-after comparison. I'd also tie every auto-applied change to a margin number before it goes live. If a headline promises free shipping and the threshold eats into your margin on small orders, a higher CTR can still lose you money. Compare profit per click, not just purchase rate.

I'd push back on the two-week before-and-after window, at least for smaller accounts. A shop doing 20 orders a month gets maybe 10 purchases in each window, so a drop from 10 to 7 is mostly noise. A payday, a holiday or a price change can move the numbers more than the new headline did, and a before-and-after comparison can't separate the headline from everything else that changed that fortnight. At low volume I'd run the window longer, or compare against an ad group that didn't get the auto-applied asset. Only after that does it make sense to argue about profit per click.