Google’s in-platform prompts look reassuringly helpful; with just a few clicks you can increase your traffic and improve your sales. They even dangle an optimisation-score uplift if you click the blue button. But those suggestions are generated by an algorithm whose primary customer is Google, not your business. Treat them as conversation starters, not commandments, and you won’t fall into the trap of overspending for quickly diminishing returns.
The Recommendations and Auto-Apply tabs surface machine-generated tactics ranging from bid-strategy swaps to budget increases. If you opt in to auto-apply, changes can go live every day without any human sign-off. The idea is that it makes PPC campaigns super easy to manage in-house, by someone without much detailed knowledge of the platform. The reality is that it takes control away from the budget holder and puts it directly into Google’s hands.
Here are a few examples we’ve seen recently where Google’s automation has caused issues:
Example 1: PMAX campaigns with no exclusions
A retailer let Google launch a PMAX campaign straight from the Recommendations tab. Because no placement or audience exclusions were set, most of the spend landed on low-intent ad inventory; revenue was flat and the retailer was (obviously) unhappy that Adwords wasn’t working for them.
Example 2: Keyword variants draining budget
Google likes to prompt you to turn on broad-match close variants to expand the reach of your campaigns. Unfortunately, we’ve seen this quickly waste budget on keywords that don’t quite make sense for the business. Broad match is great for testing, but you should be vetting for negative keywords quickly to narrow down what’s actually relevant and what actually converts.
Example 3: Automated budget uplift causing runaway spend
Google also likes to recommend that you raise your budget to avoid throttling campaigns. Just like with broad match keywords, this can quickly balloon spend, without raising leads proportionally in line. Especially for small businesses where budgets might be tight, this can cause a real problem quite quickly.
Step | What to check | Action |
---|---|---|
1. Relevance | Does it help your KPI (profit, qualified leads, LTV)? | If unclear, reject. |
2. Impact size | Use drafts/experiments to estimate lift vs. risk. | Test 50/50 for two conversion cycles. |
3. Data sufficiency | Will the change reset learning or split data pools? | Apply gradually or segment campaigns. |
4. Monitoring | Set alerts for CPA / ROAS / impression share swings. | Reverse quickly if metrics slide. |
Always eyeball them first; even “safe” fixes can clash with brand tone or regulated-industry rules.
Google’s optimisation score and recommendation prompts can make it feel like the account is underperforming, but remember that many auto-suggestions chase spend, not strategy. Treat each prompt as a hypothesis to prove or disprove with controlled testing. By adding human context and commercial common sense, you’ll keep Google Ads working for your bottom line, not the other way round.
Like what you’ve read, then why not tell others about it... they might enjoy it too
If you think Bronco has the skills to take your business forward then what are you waiting for?
Get in Touch Today!
Discussion