The Google Ads Recommendations I Push Back On (From Someone Who Used to Work There)

I spent years at Google before starting my own PPC consulting business, so I have a lot of respect for the people on the Google Ads team. Most account reps genuinely want to help you succeed, and some of what they recommend is solid advice. Having a good relationship with an ads rep can also be really helpful, even if you don’t always agree with their suggestions.

That said, I've also seen recommendations that make sense from Google's perspective but not necessarily from the advertiser's. Google reps also have growth targets. They are measured on spend, on adoption of new features, and on optimization scores, rather than on developing recommendations that are based on your specific business model, budget, or margin. Here is my take on six Google Ads recommendations that consistently come up in rep calls, and what I typically do about them.

Recommendation	My Verdict	Effort Level
Switch to Broad Match	Proceed with caution	High
Try AI Max	Run as experiment	Medium
Increase Budget	Evaluate CPA first	Low
Auto-Apply Recs	Stay opted out	Low
Performance Max (Lead Gen)	Requires offline data	High
Lower tROAS/Raise tCPA	Check margin headroom	Medium

"Switch to Broad Match"

Broad match is consistently recommended by Google, and it can work. When you have strong conversion tracking in place, a good volume of conversions, and Smart Bidding that has enough data to work with, broad match can surface relevant searches you may not have thought to target.

The problem is that the conditions that make broad match work are not in place for a lot of accounts. If you are working with a smaller budget, limited conversion data, or a narrow target audience, broad match will spend your budget across a wide range of searches and the algorithm will not have enough signal to tighten it up. Start with exact and phrase match, get your conversion tracking solid, then gradually test broad match.

Pro Tip: Only switch to broad match when you have mature conversion tracking and sufficient data.

"Try AI Max"

AI Max is Google's newer campaign feature that layers additional AI-driven targeting and creative adjustments onto Search campaigns. It is getting better, and I have seen it perform well in some accounts.

But this doesn’t mean that it will reliably work for every account. The results are still inconsistent across different account types and industries. If you are curious, run it as an experiment on a subset of your campaigns first, before rolling it out account-wide. Let the data tell you whether it is adding real value for your specific situation rather than applying it broadly on a rep's recommendation.

Pro Tip: Always run AI Max as a split-test experiment against existing campaigns.

"Increase Your Budget"

From Google's perspective, more budget means more spending. But while they unambiguously benefit, it may not be what makes most sense for your business.

Before increasing a budget, you should evaluate whether conversions are already coming in at a profitable cost. If your campaigns are converting efficiently and you are consistently hitting your budget limits, then increasing the budget makes sense. But if your CPA is already higher than you would like, or your conversion rate needs work, adding more budget is not the answer, and you should focus instead on getting these metrics right. 

Pro Tip: If your CPA is high, fix conversion metrics before increasing budgets.

"Enable Auto-Apply Recommendations"

I would be extremely cautious here. Auto-apply lets Google automatically implement recommendations in your account without you reviewing them first. Some of those recommendations are reasonable. Others will add broad match keywords, adjust bids, or expand your targeting in ways you would not have chosen yourself. Recently, I discovered during an audit that a search campaign was opted into the Display Network which generated 0 leads to the client and only increased their ad spend.

The right default for most accounts is to stay opted out of most auto-apply settings, review recommendations manually, and apply the ones that actually make sense for your goals. It is also worth checking your auto-apply settings periodically, because Google sometimes pre-checks new options when they are introduced.

Pro Tip: Keep auto-apply off. Google’s incentives don't always align with your profitability.

'"Run Performance Max for Lead Gen"

Performance Max can work for lead generation, but it requires solid offline conversion tracking. Without it, PMax optimizes toward form fills or phone calls, which might or might not make sense for your specific account. If the algorithm does not know which leads actually became customers, it will optimize for volume rather than quality, and lead quality tends to suffer.

Before running PMax for lead gen, make sure you are feeding qualified lead data or actual sales data back into Google Ads. Without that signal, you are asking the algorithm to guess, and it will guess in favor of whatever is easiest to generate.

Pro Tip: Feed PMax high-quality offline data to avoid optimizing for low-quality volume.

"Lower Your Target ROAS / Raise Your Target CPA to Capture More Traffic"

This is a frequent recommendation, framed as a way to "unlock more volume." In theory, loosening your targets gives the algorithm more room to bid on auctions it was previously skipping.

But in practice, this only makes sense if you have margin headroom to absorb a worse return. Before adjusting either target, you should know your break-even ROAS or break-even CPA. If a rep suggests lowering your target ROAS from 400% to 300%, do the math on what that means for your profitability before agreeing to it. 

Pro Tip: Know your exact break-even point before loosening targets.

Google Ads Recommendations - the Bottom Line

While any of these individual Google Ads recommendations might be appropriate in a given account, whether they actually make sense will depend on the specific context, which a Google rep typically won't have.

If you want to think through what is actually worth testing in your account, I offer a free 30-minute consultation and would be happy to take an initial look.