Not every recommendation labelled as a “best practice” in Google Ads is built with your goals in mind.
Many of them work under ideal conditions, with the right budget, clean conversion tracking, and ongoing optimization, but most businesses don’t operate in a lab.
Some of these practices can quietly erode performance if applied blindly. Others benefit Google more than they benefit your bottom line. That doesn’t mean they’re all bad, but they need context, experience, and active management.
Here are my top six “best practices” that deserve a second look and what to do instead.
1. Broad Match (in the wrong hands)
Broad match has evolved a lot in the last few years. Google has positioned it as a must-use in combination with Smart Bidding, and in the right setup, it can drive volume and surface high-converting queries you weren’t targeting directly.
But it’s also one of the easiest ways to waste budget if you don’t have the right guardrails in place. Smart Bidding only works as well as the signals it’s trained on. If you’re feeding in low-quality conversions or if your conversion tracking is messy, broad match can spiral fast.
Even seasoned advertisers have seen irrelevant traffic balloon when negatives aren’t managed aggressively or when broad match is turned on too early in a campaign lifecycle. In lead gen, this often means paying for top-of-funnel queries that will never convert.
Google claims that broad match helps you capture more conversions at lower CPAs. That might be true — but only if the rest of your account is airtight.
Actual Best Practice: Test broad match only after you’ve:
- Verified high-quality conversion tracking
- Built out a strong negative keyword list
- Stabilized Smart Bidding performance (and confirmed it’s optimizing the right actions)
Don’t default to it. Use it intentionally.
2. Automatically Applied Recommendations (aka AAR = 🤡)
Automatically Applied Recommendations allow Google to make changes to your account without your explicit review. While this feature is framed as a time-saving one, it can (and will) make significant changes that negatively impact performance.
For instance, a Reddit user shared their experience where, following a Google Ads representative’s suggestion*, they enabled certain auto-apply options. Subsequently, their search ad conversions plummeted, and customer interaction patterns changed noticeably. Despite reverting the changes and disabling AAR, the campaign’s performance took some time to recover. This shows us the potential risks associated with auto-applied changes, especially when they alter key campaign settings without your approval.
Actual Best Practice: Leave AAR turned off. Review and manually apply recommendations that align with your objectives regularly. And if your agency has this on, ask why. Then… maybe fire them.
*Google Ad Reps can be great, but please remember you’re the expert. Take their recommendations with a grain of salt.
3. The Optimization Score Obsession
Let’s clear the air, once and for all: Google’s Optimization Score is often treated like a benchmark, but it’s not. As Ginny Marvin, Google Ads Liaison, has stated repeatedly:
Optimization score is a diagnostic tool only. It does not directly affect ad rank or performance.
Despite this, many less experienced advertisers still feel pressure to push it to 100%. That usually means accepting suggestions that increase spend, expand targeting, or reduce control – suggestions that may not align with your campaigns goal in that moment.
Examples of this include:
- Swapping exact match for broad match
- Raising budgets before conversions stabilize
- Adding generic audiences or locations
If you really want to see a 100% optimization score, you can ignore the suggestions and still achieve the same result.
Actual Best Practice: Use Optimization Score to audit and review. Don’t chase 100%. Treat each recommendation as optional, because it is.
4. Going All-In on Performance Max
There’s no denying it, Performance Max (PMax) has become a huge component in many advertisers’ strategies, particularly for e-commerce and high-volume lead generation. But using it as a default campaign type without serious thought can lead to challenges, especially when it comes to insight and control.
Historically, PMax offered limited visibility into channel performance and lacked the ability to add negative keywords without contacting Google support. These limitations made it difficult to manage spend effectively or optimize based on intent.
Recent updates have thankfully addressed some of these concerns. Google has introduced channel-level performance reporting, filtered search term visibility, and the ability to add negative keywords directly within the platform. These new features will give advertisers more tools to manage their campaigns effectively.
Despite these improvements, challenges will, obviously, remain. Reporting is still developing, and asset-level transparency is limited. Also, advertisers cannot fully control how spend is distributed across channels like YouTube, Display, Shopping, and Search. This lack of control can make it harder to align performance with specific business objectives. With that said, the ability to see how spend is distributed and see conversions will help advertisers figure out what channels to divest from by pulling assets.
Mike Ryan and the great folks at SMEC analyzed over 4,000 PMax campaigns across more than 500 advertiser accounts. The report found that while PMax’s cost share peaked at just under 82% in May 2024, it has since declined by nearly 6%, showing us the shift as advertisers explore other campaign types.
The report also mentions that PMax campaigns need at least 30 monthly conversions, and ideally 60 or more, for optimal performance. Insufficient conversion volume can lead to volatility and underperformance.
Actual Best Practice: Use PMax in accounts with sufficient conversion volume to train the machine properly and with conversion-focused goals. Try not to solely rely on PMax, especially if you require precise targeting, channel-level control, or detailed performance insights.
5. “Spend Your Entire Daily Budget” – Not So Fast
One of the common assumptions in Google Ads is that you should always spend your full daily budget, or that spending more inherently leads to better performance. But this isn’t always true, especially when you’re still validating conversion quality or testing new campaigns.
What many advertisers don’t realize is that Google can exceed your set daily budget by up to 100% on high-opportunity days. This is known as overdelivery, and it’s built into the platform.
“On days when your ads are more likely to get clicks and conversions, you may spend up to twice your average daily budget,” according to Google Ads Help.
Because Google calculates spend against a monthly limit (your daily budget × 30.4), advertisers don’t need to worry about hitting their full daily spend every single day. The system is designed to shift spend across the month, not force consistency day to day. That flexibility gives you room to pull back when performance isn’t strong, without wasting your budget.
Actual Best Practice: Scale budgets based on performance, not platform pressure.
If your CPA isn’t where it needs to be, spending more won’t fix that. It tends to make it worse. Focus on conversion quality first, then increase budgets when you see consistent results.
6. Dynamic Search Ads (aka “Let Google guess”)
DSAs are marketed as an easy way to “fill in gaps” by allowing Google to automatically generate headlines and match to relevant queries based on your website content. But, their effectiveness largely depends on the nature of your business and the structure of your website.
According to Google Ads Help, DSAs are most effective for advertisers with a well-developed website or a large inventory, as they use your website content to target your ads and can help fill in the gaps in your keyword-based campaigns. They are good for capturing additional traffic and finding new serving opportunities you aren’t already targeting with keywords.
Something to note is that DSAs should be used selectively, especially for service-based or website-limited content. They can be great at finding long-tail searches or filling in the gaps with your keyword strategy, but they require monitoring and continual optimizations to ensure relevant traffic is coming in. DSAs are not meant to be the entire strategy, but rather a supplement to keyword campaigns. Use specific URLs or categories to keep DSAs focused.
In e-commerce scenarios, DSAs can sometimes duplicate queries already handled by Shopping campaigns or Performance Max, resulting in overlap without incremental value.
Actual Best Practice: Use DSAs selectively. Ensure your website has well-written HTML page titles and clearly written content about your business and industry. Regularly review search term reports and implement negative keywords to maintain relevance. Monitor performance closely to ensure DSAs are contributing positively to your campaign goals.
Final Thought
The phrase “best practice” can be misleading. These are frameworks, not rules. They require context, strategy, and regular review.
Many of Google’s promoted defaults just don’t hold up if you’re working with small budgets, limited conversion volume, or complex customer journeys. That doesn’t mean you should ignore all recommendations, but it does mean you need to be selective. Test. Validate. Push back when needed.
Because at the end of the day, your goal isn’t to run Google Ads the way Google wants you to — it’s to run them in a way that actually drives results.
