Fix Smart Bidding: Is Manual Bidding Making a Comeback?

I saw a tweet recently claiming that manual bidding was making a comeback, and I absolutely love this topic. It gets right to the heart of the constant tension between automation and strategic control in Google Ads.

Let’s be clear: Smart Bidding works great for a reason. It bids on a huge number of factors we could never manage manually, especially for Shopping Ads where we don’t even have access to the most important signal of all—the search query. That’s why it was so effective when it rolled out.

The drawback, however, is a big one: some clicks have real value even if they can’t be tracked, and Smart Bidding can’t bid for what it can’t see. So, should you switch back to manual bidding? What are the alternatives to get the best of both worlds? That’s what we’re going to explore.

The Core Problem: Smart Bidding’s Blinders

When it comes to the Smart Bidding vs. Manual Bidding debate, there are two areas of complexity that cause most of the problems.

1. Long Customer Journeys Get Ignored

Top-funnel search queries often get under-prioritized because conversions can’t be attributed back to the original keyword a user typed. These top-funnel touchpoints are critical. If a user already knows you when they search for a bottom-funnel keyword, your click-through rate will be higher, leading to more sales volume.

But sometimes that same user doesn’t come back through a Google search at all. They might come in through Meta ads, direct traffic, email, or SEO. When that happens, Smart Bidding assigns zero value to that initial top-funnel keyword. Over time, the algorithm simply stops bidding on them, and you end up with an account that just doesn’t scale. In my opinion, top-funnel keywords are essential for growing accounts, especially those with a high AOV.

2. “Worthless” Clicks Aren’t Really Worthless

Traffic has value even if we can’t see a direct conversion. If 100 clicks lead to 5 conversions, it’s easy to see the value of those 5 clicks. But does that mean the other 95 clicks that didn’t convert had zero value for your business? Of course not.

The real question becomes: at what point do you gain more by taking manual control versus staying efficient with Smart Bidding? The answer is almost never “go 100% manual.”

Why a Full Switch to Manual Bidding Usually Fails

In 90% of the accounts where we test switching to manual bidding—either for the entire account or for small parts of it—Smart Bidding still performs better when measured by Google’s tracking. And more importantly, it holds up when we verify the numbers in the back end.

Here’s an example from one of our recent experiments. We took a campaign and switched it to manual bidding to see if we could force more volume.

You can see where we made the change. We started getting more clicks and a bit more conversion value. Eventually, we were getting a lot more clicks and about 50% more conversion value.

Looks great on the surface, right? But when we look at the efficiency, the story changes completely. The return just wasn’t there. It simply wasn’t efficient to buy that much additional traffic.

Could we buy more revenue and more customers? Yes, definitely. Was it profitable? No. The truth is that outside of some very niche B2B or specialty accounts, it hardly ever makes sense to make a complete transition to manual bidding.

At the same time, for accounts in a growth stage, clicks are a KPI we’ve brought back. It’s clear there’s value in getting relevant traffic, even if direct conversions aren’t there. So, how can we implement the good parts of manual bidding to get the best of both worlds?

Wartime vs. Peacetime: Know Your Strategy

I look at accounts from two perspectives, which I categorize as “Wartime” and “Peacetime.” (I stole this from a business book, and I love how it applies to Google Ads).

• Wartime: Profitability is key, and there is no margin for error. You spend your budget only where it’s evident you get a return. This is where you let Smart Bidding be as efficient as possible.
• Peacetime: You have headroom in your profitability targets. You are looking to grow and can handle some waste. This is the time to expand beyond what looks good on paper.

When you’re in Peacetime, you can use specific tactics to boost traffic when Smart Bidding is trying to be a little too efficient.

4 Tactics to Fix Smart Bidding (Without Going Manual)

Here are the tactics we use to push for growth, ranked from least useful to my absolute favorite.

#1: Smart Bidding Exploration (My New Favorite)

The reason “lowering ROAS targets” dropped to number three on my list is this feature, which is finally out of beta. It’s called Smart Bidding Exploration, and it allows Smart Bidding to hit a 10-30% lower ROAS while exploring new search terms you aren’t currently bidding on.

While it’s not primarily meant to be a traffic-driving mechanism, that’s a nice side effect. The official explanation is that it will open up Smart Bidding to bid for less obvious search queries. The classic example I use is for running shoes. The system knows to bid on “running shoes,” but what about terms like “learning to run,” “first time runner,” or “running shoes for beginners”? These are less obvious terms that could be great to get traffic from.

Combining Smart Bidding Exploration with minimum bids can be a powerful one-two punch to ensure you’re not letting Smart Bidding become too single-minded in its focus on efficiency.

#2: Minimum Bids in Portfolios

I do believe it’s criminal that you have to use portfolio bid strategies to get access to minimum bids, but it is what it is. Minimum bids tell Smart Bidding to always bid at least X amount, no matter what your conversion data says. This is an excellent way to ensure you’re always bidding a minimum amount to get traffic.

We recently ran a test where we removed the minimum bid limit just to see what would happen. The results were immediate and dramatic. Clicks fell off a cliff, dropping from around 3,000 per day to almost half that, literally overnight. Conversions barely changed, but our traffic disappeared.

This is a perfect example of Smart Bidding not seeing the full value. This campaign was for our client’s bestsellers; I guarantee there’s value in buying those clicks, but we just can’t track it directly. The key here is that if you use minimum bids, you need to be more aggressive with negative keywords to weed out irrelevant search terms.

#3: Strategically Lowering ROAS Targets

When you lower your ROAS target, you’re telling Smart Bidding to do one of two things:

1. Increase bids for existing, converting segments (search queries, audiences, devices, etc.).
2. Find new segments to bid on because it now has more headroom in the target.

Lowering your ROAS target before a peak season or strategically throughout the year is a great way to increase traffic and give Smart Bidding the ability to find new search terms. As you can see in this example, we lowered targets and got 20% more clicks and 31% more conversions. Our ROAS went down by 13%, but I will absolutely take that trade-off for the growth we achieved right before a peak period.

#4: Manual CPC Fallback Campaigns

This works by creating duplicate campaigns (Search or Shopping) that use manual bids. Technically, if the bid in your fallback campaign is higher than what Smart Bidding is willing to pay in the primary campaign, the fallback campaign will show up instead. This means that whenever Smart Bidding isn’t bidding high enough, our fallback campaign enforces a minimum bid, ensuring we get the traffic.

We used to do this a lot, but with more options from Google, we rarely use it anymore. It’s a bit clunky but can still be a viable tool.

How to Prove This “Extra” Traffic Is Actually Working

Now that you’re driving more traffic, how can you verify it’s a good idea? The best way is with Google’s native incrementality testing. The announcement from Google Marketing Live 2024 lowered the bar for entry significantly. It used to be a $100,000/month minimum spend; now it’s down to $5,000/month.

I’m not 100% convinced a $5k test is robust, but the previous limit was prohibitive for almost everyone. Now, we can run a real test to see if we get more conversions overall from running minimum bids, lowering ROAS targets, or using exploration.

The goal is to answer one question: does driving this additional traffic lead to additional conversions we would not have gotten otherwise? On paper, the campaign without these tactics will look more efficient. An incrementality test shows you the reality of what’s happening across your entire business.

If you can’t run an incrementality test, then using a blended ROAS metric to confirm you’re still profitable overall can be a good failsafe. It’s not as precise, but it will tell you if you’re pushing for traffic at the expense of your bottom line.

[TL;DR]

• Smart Bidding’s Big Flaw: It’s highly efficient but undervalues untrackable top-funnel clicks and long customer journeys, which limits an account’s ability to scale.
• Don’t Switch to Manual: In our experience, a full switch back to manual bidding is almost always less efficient and less profitable, even if it generates more revenue.
• Use a Hybrid Approach: The best solution is to augment Smart Bidding with tactics that force it to value traffic. This works best when you’re in a “Peacetime” growth phase.
• Four Key Tactics: Your best options are lowering ROAS targets, using minimum bids in portfolio strategies, and enabling the new Smart Bidding Exploration feature.
• Verify Everything: Use Google’s new, more accessible incrementality testing or blended ROAS to prove that the extra traffic you’re buying is actually driving incremental conversions for your business.