Google Ads Performance Drop? Diagnose & Fix It Fast

Episode Number: EP025

Working Title: What to do when Google Ads stops working

Release Date: 09/04/2026

Maelie: Welcome back. Today we plan to discuss a topic that happens to marketers far more than we wish it did. Things go well. Google Ads works perfectly for you. All of a sudden, things change and lead volume drops, sometimes by as much as fifty percent or more. In today’s episode, we plan to talk through this and explain how we address and fix the issue. Louis, do you want to talk about what you do in this situation?

Louis: Yeah, sure. The first thing I always do involves isolating the exact date the change happened. You will find a specific day where you see a clear drop in the metrics. I usually look at conversion volume, conversion rate, and cost per conversion across the last ninety days. I do this just to see where I can spot a visible drop in performance. I think we should mention that these issues usually fall into three or four categories. The first category involves a change inside the ad platform. This includes user initiated actions or automatic applications Google makes. The second category involves a change on the website or landing page, like a shift in user experience. The third category involves market or political conditions, representing external factors. The fourth category involves tracking issues, like a breakdown in your tags or something functioning differently than before.

Maelie: I think we need to figure out which of those four categories applies. We talked about this separately before. A simple method exists to guide you in the right direction. If your CPC shoots up all of a sudden, this gives you a really solid indicator that you have a problem inside your ad platform. We also mentioned conversion rate. If your conversion rate plummets all of a sudden, that gives you a strong indicator that the problem lives off the ad platform. You likely have an issue with your landing page or your conversion tracking.

Louis: I think you really need to build a clear picture of what happened. You will have a before state, and you will have an after state. I typically take a window of thirty, sixty, or ninety days before the change happened. I take the same time window afterwards. I then pull the Google Ads data into Looker Studio and start to dissect where the change originates. I look for specific patterns. Did a specific campaign stop delivering? Did specific keywords stop working? Did specific search terms drop off? You can also look at the change history if you suspect a platform issue. You can check if anyone changed anything around the drop date. This helps you picture what potentially caused the dip.

Maelie: Landing pages become tricky depending on how much control you hold. If you do not create the landing pages or handle the web development, you need to bring other people into the conversation. You must ask what changes they made. You should build a suspect list of recent changes. You can then identify the most likely culprits and fix them. You also have the option to roll back the clock and restore the previous version if possible.

Louis: I think you raise a really good point. We talked about external factors as potential causes. Over the last couple of years, Google brought in Consent Mode. We talked about it a few times on the pod. You might get the same number of leads, but you only measure a much smaller portion of them. That represents an external factor where your ads look worse even though they perform exactly the same. Google also increases CPCs sometimes, so you pay more for the same amount of clicks. External factors make diagnosis difficult. Browsers change technology to restrict tracking. Market conditions shift. Seasonality changes. Generating leads naturally costs less in a peak season than in a shoulder season. The political environment influences performance too. In those situations, you cannot roll back the clock. You need to accept a new normal. Past data loses its usefulness as a benchmark. You must set new benchmarks moving forward. You cannot look back on previous months and claim you underperformed because you now operate in a state of new normal.

Maelie: Following up on what Louis said regarding market changes, you must look at your ads and landing pages first. Do not jump straight to market factors. The market represents an irreversible state. You do not want to default to that excuse. Make sure you rule out the ad platforms and the landing pages first. Once you rule those out, you can conclude an environmental or market factor caused the issue. You can then pinpoint the specific market change that suddenly made your lead volume drop.

Louis: You make a really important point. Marketers find it too easy to blame external factors. External factors do not always cause the drop. You have to eliminate all other potential causes first.

Maelie: I think we should mention some real examples. Three examples come to mind. The first one happened when leads literally dropped by fifty percent month on month. The conversion rate caused the drop, so it represented a website issue.

Louis: In that instance, the client operated a software company. We isolated the day the problem started. We used the Wayback Machine to see what the page looked like before and after. We found the problem on their integrations page. They tiled all their software integrations on a new page. All the tiles featured clickthrough links. They changed the buttons to make them prettier, more graphic, and more subtle. As a result, people missed the fact that they could click the product tiles. We reverted the design back to highly visible and obvious buttons. That simple change brought the conversion rate right back because people resumed clicking the tiles and converting on the next page.

Maelie: You should remain aware of another scenario. Quality sometimes takes a sudden downturn. We saw this with one client. They still received tons of leads, but the quality suddenly shot through the floor. Performance Max previously performed very well, but it suddenly found a new group of people. The algorithm optimised heavily into this new group, generating poor quality leads. We made several changes to bring Performance Max back in line. Do you want to talk a little bit about what you did there to improve the quality, Louis? I know people experience quality issues with Performance Max often. Any light you shed on this will prove really useful.

Louis: Performance Max acts like a self-perpetuating machine. If it sees a good volume of conversions, it acts like a terrier and chases that volume. You want to feed it the highest quality conversions possible to point it in the right direction. Secondly, do not rely entirely on Performance Max. Always make sure you run Search campaigns alongside it. You also have a manual option, acting like a search term report for Performance Max. You pull up a placement report inside the platform. You look at all the poor placements and exclude as many as possible.

Maelie: You encounter thousands of placements, so you cannot look at all of them. You can take some great shortcuts. For example, you can exclude specific country domains like dot io, dot pl, or dot xyz if you do not want your ads featured there. You can create a script to automatically list these and exclude them. You also look at your top placements. You check which ones feature heavily. You click into them to see what they look like. If they clearly look wrong, you exclude them. Exclude those children’s games because they come through too often in Performance Max.

Louis: I promised three examples. The third one comes to mind when everything goes well for a client. To react to good performance, they pushed their budget way up. That makes sense intuitively. You scale the budget up when things go well. However, you always risk scaling the budget up too fast. You spend five grand one month, and you feel really happy. On day one of the next month, you increase the budget to ten grand. You make a massive change. Google understands the parameters in a fundamentally different way. It accesses new audiences and treats the campaign very differently. They should have increased the budget twenty percent a day until they reached double the budget. To fix the break, we dropped the budget back down. We moved the bid strategy from Target CPA back to Maximize Conversions for a while. We made the strategy slightly less aggressive so Google could find its bearings again. When it started generating conversions at a decent rate again, we restored the Target CPA bid strategy.

Maelie: You must bear in mind that you will encounter a hockey stick effect when you scale. At a certain point, your cost per conversion simply rises and rises.

Louis: We talked a few times on previous episodes about using a target CPA you feel comfortable with. You work from there, rather than trying to get a better cost per lead every single month. If you lack a max CPA, your cost per lead will increase as soon as you add more budget. You do not want to end up in scaling jail.

Maelie: That feels like a good place to do a quick recap. First, things change on a specific date. You want to find the exact date your leads fell off a cliff. Next, you decide whether an ad platform issue, a website issue, an external factor, or a conversion tracking issue caused the drop.

Louis: Two metrics point you in the right direction. If your CPC increased, you probably have an ad platform issue. If your conversion rate dropped, you need to check if anything changed on the website. Knowing which metric affected the campaigns proves really useful. It lets you know whether you can reset back to the before state or whether you must accept a new normal moving forward.

Maelie: If you enjoyed this episode, please subscribe or leave us a review. Every single review means the world to us and helps us grow. If a topic nags at you, please head to webmarketeruk.com/topic and let us know for a future episode. Thanks for listening, and we will catch you on the next one.