Google promises better results with just one click—“Apply All Recommendations.” But what if those recommendations aren’t actually in your best interest? At 97th Floor, we’ve seen firsthand how blindly following Google's automated suggestions can drain budgets and derail strategy. So we put it to the test. In a head-to-head experiment, we pitted Google’s recommended setup against our tailored, expert-built campaigns. The results were eye-opening—and could change the way you manage your ad spend.

The Experiment

We tested campaigns with three of our clients to compare the effectiveness of Google’s automated recommendations with the hands-on approach taken by our experienced advertising team at 97th Floor. Each campaign was structured to align with specific goals, but the key difference was in how they were managed:

  1. B2C Ecommerce Company in the Health & Wellness Industry: This advertiser launched two campaigns—one using Google’s recommended setup and the other following our strategy, with the aim of increasing website traffic and conversions.
  2. B2B SaaS Company in the Technology Sector: A new campaign focused on branded keywords was set up using Google's recommendations and compared against an existing campaign managed by 97th Floor.
  3. B2B Digital Accessibility Service Provider: A new campaign targeting non-branded keywords was launched following Google’s setup, and its performance was compared with a similar campaign managed by our team.

The Execution

Google97th Floor
ObjectiveSelect oneSelect one
Conversion actionsAuto-selectManually select
Campaign typeSearchSearch
How you reach goalsVisit websiteVisit website
BiddingStart with conversionsStart with clicks
Max CPC limitDon’t start with oneStart with one depending on client; not too restrictive (e.g., under $10)
NetworksAllow Search partners and Google Display NetworkTurn Google Display Network off; test Search partners if...
LocationSet a location as presence or interestSet a location as presence only
Audience SegmentsObservational targetingObservational targeting
Broad matchOnOff
Automatically created assetsOnOff
Ad rotationOptimize for best performanceOptimize for best performance
Ad scheduleNoneAdd if needed
Campaign creationGo through Google's processGo through 97th Floor's process
Ad copyUse Google's suggestionsWritten by expert content marketer
BudgetEqual budgetEqual budget
OptimizationsTurn on auto apply recommendations for this campaignOptimize manually
Yellow highlight indicates an option for which the 97th Floor team did not use Google's recommendation.

Constants

Here are the settings we kept the same between the control and test group. These are Google’s recommendations that we agree with.

Google97th Floor
ObjectiveSelect oneSelect one
Campaign typeSearchSearch
How you reach goalsVisit websiteVisit website
Audience SegmentsObservational targetingObservational targeting
Ad rotationOptimize for best performanceOptimize for best performance
BudgetEqual budgetEqual budget

Changes

Conversions

When you feed a smart bidding system bad signals, you get bad results. We focus on signal quality to guide the algorithm toward actual business growth — not just form fills or button clicks that look good on paper.

Google97th Floor
Conversion actionsAuto-selectManually select

Bidding

Smart bidding needs data to perform well. Jumping straight to "Maximize Conversions" without historical data is like trying to sprint before you've learned to walk. Our approach allows us to build a strong foundation — minimizing waste while setting the campaigns up for smarter automation later.

Without a CPC cap, Google can overspend on low-quality traffic. We applied reasonable limits to protect budget efficiency.

Google97th Floor
BiddingStart with conversionsStart with clicks
Max CPC limitDon’t start with oneStart with one depending on client; not too restrictive (e.g., under $10)

When and Where Ads Show

The Display Network may offer more impressions, but not always better ones. Especially early on, we want every click to count — and that means keeping the focus tight on search intent, not expanding blindly into display.

Advertising to someone thinking about your market is not the same as advertising to someone in your market. Having your location set as presence or interest allows Google to waste your dollars on the wrong users.

Google's default runs ads 24/7. We customized schedules to align with peak engagement and business hours when necessary.

Google97th Floor
NetworksAllow Search partners and Google Display NetworkTurn Google Display Network off; test Search partners if...
LocationSet a location as presence or interestSet a location as presence only
Ad scheduleNoneAdd if needed

Keyword Targeting

Broad match can open the floodgates — sometimes usefully, sometimes recklessly. We prefer a controlled expansion. Typically, we prioritize stricter match types, like phrase or exact. We use broad match strategically, mostly on longer tailed terms since the intent is more spelled out. It’s important to test broader match types strategically—you don’t want to waste ad spend but you do want to take into account that 15% of searches every day are new. People could be in the market for your product or solution but type it in differently than your keywords. Broad match keywords are a good way to mine for new keyword ideas.

Google97th Floor
Broad matchOnOff

Assets

Auto-generated ads often sound robotic and generic, or can state things that are outright wrong. We believe brand voice is too important to automate away — especially when you have just seconds to grab a user’s attention and build trust.

Google97th Floor
Automatically created assetsOnOff

The Results

In every case, the campaigns managed by 97th Floor outperformed those set up using Google’s automated recommendations. Here's a breakdown of the results:

  • Impressions vs. Clicks: While Google’s campaigns generated more impressions, they consistently resulted in fewer clicks. In contrast, 97th Floor’s campaigns focused on quality over quantity, yielding fewer impressions but a significantly higher click-through rate (CTR). This demonstrates the importance of targeting the right audience with the right message rather than casting a wide net.
  • Cost Per Click (CPC): Although the cost per click was higher for 97F’s campaigns, this investment paid off. Our approach led to more conversions, proving that quality clicks are more valuable than volume alone. Ultimately, our campaigns drove conversions at a lower overall cost.
  • Conversions: 97F’s strategy proved most effective with the B2B digital accessibility service provider. While Google’s campaign produced zero leads, our approach generated 45 leads at an average cost per lead of $13.30.
  • Campaign Findings: For the B2B SaaS company in the technology sector, Google’s campaign not only missed the mark on branded keywords but also resulted in poor lead quality, with most being spam. Furthermore, it had less than 10% search impression share, with 65% of this loss attributed to poor rank. In comparison, our campaign accurately targeted the intended audience, leading to higher quality interactions.

Why Agencies Still Matter

The data speaks for itself—relying solely on Google’s recommendations can leave performance on the table. While Google’s automation offers convenience, it lacks the precision and strategic insight required to maximize advertising budgets effectively. 97th Floor’s campaigns delivered better results, proving that expertise and human intervention are essential for optimal ad performance.
If you're relying entirely on Google’s automation, you could be spending more without seeing the results you deserve. At 97th Floor, we offer expert campaign management that outperforms Google’s one-size-fits-all approach. Contact us today to see how a professionally managed campaign can transform your advertising results.

Ready to grow?

See how a professionally-managed campaign can transform your advertising results.