Menu
Updated: 04/03/2024
Google ads has changed drastically since 2021, keeping on top of all of the changes and why google ads don’t work now even though they’ve been working perfectly for years before is difficult. I created this write-up to help those who are trying to manage their own Google ads account or want to learn how to and know the most common issues that occur.
It’s getting more difficult for individuals to manage their own accounts due to the lack of information and courses to properly manage the account. Even taking Google ads Certification won’t help much as it’s only covering very basic features and scenarios, even the specific certifications for shopping and display, etc. Have a lot of General Google ads questions mixed in.
The cost of having an expert/agency manage your account does include them keeping on top of all these issues and changes on your behalf. Most agencies or freelancers manage multiple accounts and can find any new changes or new issues faster than individuals managing a single campaign. This allows them to prevent these issues from happening or fix the issue faster as they know what to do.
Broad match keywords can be useful when done correctly and are best used as testing via experiments since they can match searches to any word in a keyword bringing in tons of irrelevant and top funnel searches, which is why you need a lot of negatives. The push from Google is that this match type can improve conversion rate, cost, etc. But from what we’ve seen this isn’t always the case, most of our campaigns have beaten the broad match .
Even phrase and exact match keywords sometimes stray heavily away from the keyword you provided.
With Google’s “Close match variant,” they can remove words or add words that can change the meaning of the original keyword and irrelevant target searches.
Google’s AI has not figured out the context or Intent, which means we need to constantly review searches to make sure your money is being spent on relevant searches.
Running Pure broad match for, say a service industry like home services would require tens of thousands of negative keywords in order to be pretty sure the searches coming in are relevant and not top-of-funnel people looking for DIY or how to, or even just toying with the idea of possibly getting this work done this year or maybe in a few years.
Having pure broad or phrase match keywords with a handful of negatives, one hundred or less, has a high chance of bringing in irrelevant and very top funnel searches that rarely end in conversions for that month or even quarter.
To run successful pure broad match keywords you typically need in the hundreds of thousands of negatives in order to reign in the straying of Googles machine learning and only focusing on the actual relevant phrases for your industry/niche.
Even with running regular shopping ads having a very niche product can be incredibly difficult to narrow down as it seems Google always finds a way to match to something irrelevant. We ran shopping ads for a very niche ecomm client selling satanic themed decor and we had 25,000 negative keywords with a mixture or broad, phrase, and exact matches to stop showing for generic decor searches as they aren’t our target audience. Every day we were adding hundreds of negatives and it got to a point where it wasn’t worth continuing for the client because no matter how many negatives we added we kept getting the same types of searches coming through. Maybe at 250,000 we may have finally niched it down.
We have audited dozens, and dozens of SKAG-built (Single Keyword Ad Group) accounts due to performance issues, and they always fall under the 80/20 rule, where 20% of the campaigns or ad groups, or even keywords, bring in 80% of the conversions. We’ve had accounts with 30 campaigns and 180 ad groups with thousands of keywords, and even then, 5 campaigns had all of the conversions.
The two biggest reasons this fails are, as mentioned above, Google is veering off of keywords and matching to what it thinks you’re targeting which causes a lot of competing ad groups, second is that they’re built right out of the gate with no data to back them up. We’ve seen this with other large agencies who have seen these work well with their other clients, and they do, but overbuilding a new account is just a waste of time and money if they charge monthly since it takes a lot of time to optimize a failing SKAG build.
STAG builds are what has worked in the past and still today, even with all of Google’s changes, STAG builds (Single Theme Ad Group) group all similar keywords into a single ad group. That way, even if Google starts veering off, it’s at least staying within that ad group’s theme. STAG builds are easier to build as most just mimic the layout of the website. This is the same for E-commerce and Service or other industries. Usually, only a single campaign is needed to start, and then each ad group is a service or shop category. If your services or categories have further breakdowns, you can then break out each service into its own campaign, and each sub-category would be the ad group. Doing the later, you need to have a big budget and enough sub-categories for it to make sense in breaking down that much.
Google has pushed these smart search campaigns (and now Performance Max) as an easy setup, fully automated campaign to get results without really needing to manage the account. They are easy to set up and can bring in leads or sales; however, they’re still not as good as having an expert or agency set up a well-built campaign.
We can always outperform a smart search campaign and, a lot of times, a performance max campaign. We’ve even used these as our salesman with weary clients who didn’t want to spend for an agency to manage their account; we would tell them they could test building the smart campaign themselves and let it run for a few months, and if the performance isn’t up to their standards then come back and see us. We always heard back.
We also had a proof of concept come back our way with an old client who hadn’t touched the campaign all year (9 months in 2022) and wanted us to review and optimize the account, then hand it back. All year with no changes to any of the campaigns (our search campaign, a display campaign, and 2 smart search campaigns).
Ours outperformed all of them in a number of conversions and conversion rates, even though the smart campaigns were continually optimizing themselves.
If you’re running regular RSA (responsive search ads), which are sending traffic to your website, you need to make sure you have conversion tracking set up on your website; otherwise, you’re missing out on a large portion of conversion data to know your ads are pulling their weight. With most service-based companies, a lot of times, we see a pretty even split of leads coming from calls directly from the ads and calls or form submissions on the website.
Just tracking calls from ads is beneficial, but also tracking calls from the website and form submissions or newsletter signups can also be just as beneficial to know, especially if you have email marketing.
It’s nearly impossible to optimize a poor-performing campaign with 0 conversion data, even if it’s been running for a year or two. The data is irrelevant as we cannot verify if the traffic and clicks benefited you, which specific clicks led to a new customer, or how many clicks it took to get a conversion to make sure we’re hitting a reasonable cost per conversion.
Learning a bit about setting up conversion tracking on the website is very beneficial. Using Google’s tag manager is the best way to go, as you can duplicate nearly every conversion you set up for Google ads and send it over to your GA4 analytics. Doing this will allow you to track which sources outside of Google ads are bringing in the most leads.
We’ve been noticing over the past few months, Q3/4 2023 and Q1 2024, that the search partners network has been inundating some of our accounts with spam. This included spam submissions, calls, and showing up in excluded or non-excludable countries.
We’ve since removed this from all of our client accounts in the search campaigns; Pmax, on the other hand, still uses it, we believe, and unfortunately, we haven’t found a way to exclude it yet.
I’ve shared an example from 1 clients local Pmax, we’re pretty thorough with adding excluded locations (image 1), but as you can see in the matched locations we were bombarded by DR traffic which was excluded but also Iran which we cannot exclude in Google ads. The only explanation is the search partners network had different rules that Google doesn’t check. Also you can see we only had 3% of the total clicks and interactions refunded as invalid.
This was one I thought was a bug since Q3 2023 as it only happened once across all of our accounts, but in Q4 and Q1 was happening more often. We were seeing the error that we’ve exceeded the limit for exclusions for locations, this time at least we got a number of 121 vs no mention of a number previously.
This is very concerning for us as our list of international countries alone is 195, then if we include other states for a local business we’ll add another 49, lastly if they are a small local company only wanting to focus a few counties we’ll exclude all other counties in that state, which for texas adds another 250 alone.
We don’t want to do this but have to because the AI that matches a location doesn’t work and ignores targeting at times. Example being we have a client targeting a good sized city, we excluded just a ring of counties around our target area, and sure enough they get a call from a guy right outside our exclusion ring over 100 miles south.
The response from our support rep at the time for this error was “This feature is to help elevate bloat in Google ads by adding superfluous negatives when targeting is all that should matter. That being said, if a significant amount of traffic is coming from outside the targeted area, that should be grounds for a refund.”
I don’t hold my breathe for a refund from Google as past experience has shown us they don’t give those out without some struggle, even with solid evidence. I’m hoping this is only temporary but this can be another factor as to why things aren’t working if the AI is wondering away from our target area.
The standard progress for bidding is Manual CPC > Enhanced Manual CPC > Max Conversions/Max Conversion Value > Max Conversions CPA/Max Conversions ROAS. Where Target CPA and Target ROAS are supposed to be the end for bidding strategies.
In our experience, having this linear progression is never the case, yes they all work well, but we have also seen them all perform great and stable then tank in conversions or performance, causing us to need to shift to another strategy (mostly back to manual CPC) to re-stabilize and then move back to automated again, a lot of times though we just stay in manual.
Automated bidding is beneficial as it can work more efficiently than a skilled manager and a bit better at times, but even then, we’ve found all of these have wasted spend areas, so no option is ever perfect. With the changes, Google is making, they always recommend waiting 2 weeks after any major change to allow the bidding strategies to “re-learn” this is very difficult to just wait, especially during peak seasons. We’ve been moving to Manual a lot more as we can pivot a lot faster than these automated bidding strategies, whether we’re adjusting budgets a lot, shifting focuses to specific services, or focusing most of the spending on the highest converting keywords instead of spreading it out more like the auto bidding likes to do.
All bidding strategies, even manual, are susceptible to Google doubling the daily budget or even keyword bids at any point during the month, sometimes for multiple days in a row. This is supposed to, according to them, even out over the month, but we’ve seen campaigns traffic die out for days because their “AI” overspent early on and wasted a ton of budget and needs to recoup to not overspend. We’ve seen them double our daily budget for multiple days as well, or have one week send nearly double the weekly budget and the following week underspend by nearly a third, potentially losing a lot of conversions, compared to if they just spent the budget that we set.
We always recommend checking your spending on a weekly or bi-weekly rate to make sure you’re on track for the month. There’s no way to prevent this, as we’ve seen it happen in campaigns on manual CPC with enhanced turned off. If your campaign is just showing 0s for a few days, check the prior spend as this could be the culprit.
This is where a lot of issues are caused, and even some agencies will miss these, not that they’re bad at their job, but because Google keeps making changes and doesn’t notify anyone.
You want to make sure you select under Target “Presence: People in or regularly in your targeted locations” and under Exclude “Presence or interest: People in, regularly in, or who’ve shown interest in your excluded locations.”
However, just selecting these and not having excluded locations still won’t keep your ads only showing within your target area; you will want to add exclusions for every international country and outside states (if you’re a local company) for these settings to block out 98% of outside traffic. This should also exclude a lot of searches from within your target area looking for your service in a different state.
You may still show up for some due to the “regularly in” portion, especially if you’re near a state border.
I believe this is an issue with some of the platforms not being able to block out everyone and the fact Google’s AI seems to veer off ideal targeting.
The issue again is a context/intent that Google’s AI doesn’t grasp; when we ran ads for a middle school open house and had 18-24-year-olds and non-parents excluded, we didn’t realize that the optimized targeting would then show ads to these excluded areas, rather than expanding our reach within our current targeting.
Google support said the AI could only do this if it felt that those excluded areas could convert, but logical thinking knows this will never be the case in this specific scenario and is another reason to never fully trust that the machine learning is actually showing your ads to optimal targets and to always review what it’s doing.
This will save a lot of headaches trying to figure out what changed and come to find out an auto-apply was pushed through. They say they’ll email a suggestion anytime it is about to apply, but with over a hundred of accounts linked to our manager account, we’ve never received this.
Lastly, every account we’ve ever been in has always received the recommendation to increase the budget; regardless of how much we were spending, you don’t have to do this, and it won’t guarantee any more conversions than you’re getting now. 
Under account-level asset settings, there is a new automated assets section that will auto-create headlines, site links, and various other aspects of your ads. If you want control of your ad copy and message, you will want to disable these settings.
Always check in on your change history tab, as this can bring to light reasons for performance drops. This can also show any auto-apply recommendations Google may have pushed through without you knowing.
You can also see if you’re making changes enough or not at all, typically weekly adjustments are needed but at least once a month can also work depending on the stability of the account; even weekly adjustments shouldn’t be making changes to the same things each week it should be broken out over the course of the month in order to gather enough data to make informed decisions.
On the flip side, you should only be making daily or changes every other day on a brand new campaign or build, and then once stable, move to weekly checks; for very large accounts being in the account, more often may be required depending on the situation, but even then it’ll be focusing on a few campaigns at a time and not all campaigns, so it’s still spread out.
I’ll be upfront when I say the account specialists from Google aren’t bad and do genuinely want to see you succeed. But there are still quite a bit who are not as experienced and will just be reading through a support document rather than listening to your actual concerns and walking through what is best for your account and business goals.
The biggest red flag I found, and I’ve talked to A LOT of Google specialists over the years, with some great ones who genuinely cared and helped get our client account performing again to our expectations and strategy.
The ones I almost always ignore are the ones who either don’t act like they’re truly listening to you and are reading through a script just to get through it, they only focus on the recommendations tab, or they say the issues are only caused by your match types as they are all exact and not broad, or they just tell you your bidding strategy is wrong. If they reach out, you can have them email you their list of optimizations you should take and see if they fall into the above. If so, you don’t need to waste your time. These specialists also change every quarter for everyone’s account, so if you have a great rep, it’s usually short-lived.
There is a slew of reasons an account is underperforming, as you can see from this post, so just narrowing it down to a few common areas isn’t always accurate; we have also cleaned up a lot of failed accounts from business owners who DIY managed their account and was making changes based on the Google ads specialists recommendations.
Don’t feel obligated to make any changes they recommend; I tell them all the time I won’t make that change because it isn’t beneficial to our strategy or goals. You can say no and dismiss recommendations. I’ve had one rep tell me to increase a budget by $1/day to see if it improves the conversion rate, that campaign was already at $31/day and on target CPA bidding, so $1 more wouldn’t have any noticeable effect. They also wanted me to switch away from target CPA to max conversion value just to see if we get better results. We were already converting at a steady 15% rate, and changing the bidding strategy just for fun makes no sense.
The best course of action if you’re truly unsure is to have them write down all of their suggestions and email them to you, then reach out to a Google ads agency that offers free consultations and or a no-charge audit, like ours and go over the recommendations list with them for an unbiased second opinion.
Something we’ve been noticing ourselves that support isn’t aware of is that Google filters what communication methods are available to you based on your account. When we reach out to Google support, often times, I put our Manager’s account ID in the form field because I have a lot more options available than if I put our client’s account ID. The “Call Us” option is deceiving as it gives you the phone number to call, however, you have to enter the Google account ID and at that point for smaller accounts it’ll say “sorry call support is unavailable due to Covid-19” When I call back and use my MCC account ID I get right through (and then told they’ll take my question offline and email me).
Another reason is with so many accounts under our management we get contacts from a lot of support reps, we can weed through these to find the ones who have more expertise and will actually help. The benefits of this is that we will then either have them handle the clients account, or we can get direct contact to the support rep handling the account already and bypass the generic support reps.
Request Your Free Consultation
Please find a time below, call or text us anytime for your free consultation. You can also use the “Schedule A Call” button to skip a step and grab a spot on our Calendly calendar.
Office hours are currently Monday – Friday 09:00a – 06:00p CST
We look forward to speaking with you!
~MDM PPC