Solved: Limiting Spend in Google Ads Max Conv Value Campaigns

Hi there,

Thanks for your enquiry. I can totally understand the frustration. Seeing spend go wild while results are all over the place is enough to make anyone want to throw their laptop out the window, especially when you're trying to work next to a sleeping kid. It's a really common problem with automated bidding strategies if you don't know exactly how they operate under the hood.

The good news is you can get this under control. The issue isn’t so much that the campaign type is broken, but more about how you're managing it and how the account is structured. The constant tinkering is probably doing more harm than good. I'll give you some initial thoughts on how we'd approach this to get you some stability back.

We'll need to look at why 'Maximize Conversion Value' is so volatile...

First thing to get your head around is that Google's "daily" budget isn't really a daily budget. It's a monthly budget divided by 30.4. Google gives itself permission to spend up to double your set daily budget on any given day if it thinks there's a good oppurtunity to get conversions. It just promises that over the course of a month, it won't exceed your daily budget x 30.4. This is why you're seeing it spend 2-3 times your daily limit. The 80% CPC increase you mentioned is probably a big factor here – competition has ramped up, and Google is bidding higher to chase the conversions it needs to hit your tROAS, burning through the budget faster on days it sees a chance.

The visualisation below should make this clearer. It shows how even with a fixed daily budget, your actual spend can be all over the place from one day to the next.

Now, the second and most important point: every single time you adjust the tROAS target, you're essentially hitting a reset button on the algorithm's learning phase. You mentioned matching Google's suggestions every 3-4 days. This is the root cause of your wild performance swings. The algorithm needs a stable period of time, usually 1-2 weeks minimum, with consistent data to properly learn who your best customers are and when to bid for them. By changing the target constantly, you're not letting it get out of "learning mode". That day you got 5x the orders? The algorithm found a rich vein of buyers, went all in, and you got a great result. The next day, that vein was tapped out, and because it's still in a chaotic learning state, it couldn't replicate the performance and your results crashed. You've got to give it time to stabalise.

I'd say you need to regain control by setting stable targets...

So, what should you do? First, stop touching the tROAS target. Pick a realistic target based on your historical data (what's a profitable ROAS for you?) and commit to leaving it alone for at least two, preferably three, weeks. Let the algorithm do its job. I know it's tempting to react to a bad day, but you have to look at performance over a longer window, like 7 or 14 days.

To set a realistic tROAS, you need to know your numbers inside and out. What's your break-even point? What profit margin do you need? Your tROAS should be based on that, not just on Google's suggestions which are often designed to get you to spend more. Knowing your LTV (Lifetime Value) is even better. I remember one eCommerce client who sold subscription boxes; we calculated their LTV and realised they could afford a much higher acquisition cost than they thought. This allowed us to set a more aggressive but stable tROAS target and achieve a 1000% Return On Ad Spend because we weren't scared off by a high initial CPA. We knew each customer was worth much more in the long run.

Here's a little calculator to help you work backwards from your ROAS target to figure out the absolute maximum you can afford to pay for a single purchase (your CPA). This should be your North Star metric. If your CPA is consistently below this number over a 14-day period, your campaign is profitable. If it's above, then you might have a problem that needs investigation.

You probably should restructure your campaigns for better control...

Relying on one big campaign for all your keywords is another reason you're feeling a lack of control. It's like having all your eggs in one basket. The performance of your best keywords is getting dragged down or inflated by the performance of your average or poor keywords. The algorithm is trying to balance everything out, which leads to this unpredictability.

What we'd almost certainly recommend is a campaign restructure. You mentioned "some of our big KW". These are your crown jewels. They should be treated as such. We would pull these top 1-3 converting keywords out and put them into their own, separate campaign. This is often called a "Top Performers" or "Alpha" campaign.

Here’s how this works in practice:

• Top Performers Campaign: This campaign contains only your proven, highest-volume, highest-converting keywords. You give it a dedicated, healthy budget and a slightly more ambitious tROAS target because you know these keywords convert well. Your ad copy can be hyper-specific to these terms. This campaign becomes your reliable, stable engine for revenue.
• General Campaign: This campaign houses all your other, broader keywords. It can have a more conservative tROAS target and a more flexible budget. Its job is to capture a wider range of traffic and, crucially, to identify new "big KWs". When a keyword in this campaign starts performing exceptionally well over a few weeks, you "graduate" it to the Top Performers campaign.
• Brand Campaign: If you're not already, you should have a separate campaign just for people searching for your brand name. This should be a low-cost, high-ROAS campaign that simply protects your brand from competitors bidding on your name.

This structure gives you far more granular control. If your General campaign is having a bad week, it doesn't drain the budget from your most reliable keywords. You can adjust budgets and targets at a campaign level with much more confidence. It does take more work to set up, but it pays off massively in stability and scalability. We've used this exact approach for eCommerce clients and it consistently leads to more predictable returns and better overall performance. Isolating your top performers protects their budget and allows you to scale what you know works, rather than having its performance diluted by less predictable, broader terms.

You'll need to look at your data differently...

The final peice of advice is to change how you look at the data. The day-to-day rollercoaster is a distraction. You need to zoom out. Start looking at performance in 7-day or 14-day rolling windows. Is your ROAS for the last 7 days meeting your target? Is your CPA for the last 14 days below your maximum affordable CPA? These are the questions that matter.

Don't get spooked by one bad day. Equally, don't get over-excited by one amazing day. The algorithm is playing a long game, and you need to as well. The goal isn't to have a perfect ROAS every single day, it's to have a profitable ROAS over the course of a month, and then a quarter. When you adopt this mindset, you'll stop making reactive changes that sabotage the campaign's long-term preformance.

This is the main advice I have for you:

I know this is a lot to take in, so I've put the main recomendations into a table for you. This is the step-by-step plan I'd suggest you follow to get things back on track.

Managing automated campaigns like this is less about constant adjustment and more about building a solid structure and then having the discipline to let the system work. It can be difficult to do this yourself, especially when it's your own money on the line and you have the pressure of running the rest of the business.

Working with an expert can help provide that objective viewpoint and implement the more complex structures needed for stable scaling. We spend all day inside accounts like yours, so we've seen these patterns hundreds of times and know how to fix them.

If you'd like to go through your account in more detail, we offer a free, no-obligation initial consultation where we can review your strategy and give you some more tailored advice. It might be helpful to have a second pair of expert eyes on it.

Hope this helps!

Regards,

Team @ Lukas Holschuh