The Four Most Common (& Costly) Google Ads Mistakes

An Expert’s Opinion From 10+ Years “In The Weeds”

Here’s a good analogy: Landing pages are like car exteriors, whereas keyword selection and account structure are like the car’s internal components.

A mechanically flawless vehicle will have trouble selling if the exterior is dinged up and painted some hideous color.

Similarly, a bad landing page, no matter how stellar your keyword selection and account structure, will be a major drag on performance, leading to hemorrhaging of budget.

Conversely, an amazing landing page with supreme attention paid to conversion rate optimization (i.e. a Ferrari-like exterior) can help to make up for a poor account structure and sub-par keyword selection.

The notion that landing page design (and conversion rate optimization) is just as crucial as keyword selection and account structure is especially true these days given the increasing popularity of Google’s automated bidding strategies and its expansion of close match keyword variants, which allow Google more and more control over how your budget is being spent.

Pairing a world-class account structure with a high-converting landing page is the most important recipe for success. But too often landing page testing and optimization get overlooked.

Why?

1. Corporate bureaucracy

This is not an excuse (even though we see it all the time). A corporate agenda to actively disengage from experimentation and iterative improvement leaves the doors wide open for your competitors to sweep your potential customers away from you – all because of red tape. Not experimenting with alternative landing page layouts/designs that might stray from your brand guidelines/design aesthetic is very shortsighted.

2. You just don’t know better

Now you do.

3. Limited resources

If by resources you mean lack of development capabilities, there are loads of magnificently simple landing page tool out there. Here are a few of the most popular ones: Unbounce, Leadpages, Instapage.

If by resources you mean lack of budget, the above tools cost $50-200 a month. For some companies, your improved CPL will pay for the cost of the tool in one or two conversions.

4. You don’t know how to improve it

This is the hard part, as conversion rate optimization is a nuanced skill that takes a lot of experimentation and iteration. You have to be OK with failing in order to get better. The key things to keep in mind are:

• Keep it simple. Leave your corporate marketing jargon at the door. A high schooler needs to be able to understand your main value props in 3 minutes and then regurgitate them to you.
• Take advantage of your area above the fold. Most visitors don’t make it past the content they initially land on. Say what you need to say ASAP, in a compelling way, and leave the details for below the fold.
• Urgency, scarcity, and social proof. These are the three things that move the needle.

We suggest testing 3 different landing page designs from the start, allowing you to quickly identify a general design that performs best. From there, additional iterations can (and should) be made to further fine tune performance and get the most out of your marketing budget.

See how simple and to the point this Landing Page is:

While landing page testing (or lack thereof) is the mistake that can move the needle the most for properly setup campaigns, the improper set-up and subsequent fake/junk leads that result are the most common mistake we see.

The ramifications of fake leads are huge!

They have a terrible negative cascading effect on your program, not only clogging up your CRM, but these fake leads actually TRAIN google to find more of them (because of auto-bidding strategies)!

So not only are you wasting money on them, but you’re telling Google to find you MORE of them!

No más!

Below we’ll explore a five-part strategy for eliminating fake leads.

The first two parts are technical/may require a developer’s help, whereas the 3^rd, 4^th, and 5^th parts are specific to how you’ve set up Google Ads.

All five parts should be followed for best results.

Part 1: Set up ReCAPTCHA and a honeypot

Most people know of ReCAPTCHA – Google’s tool for making sure you’re not a robot. There are three main variations of ReCAPTCHA: v2 (“I’m not a robot” box), v2 (invisible), and v3 (score-based).

There’s no right answer for which to use. The only right answer is to definitely use one of them!

You can learn more about ReCAPTCHA here, and set up ReCAPTCHA for you website here.

The second part of this step is setting up a “honeypot”.

A honeypot is a hidden/invisible form field on your form. It is impossible for real users to fill out (because it’s hidden), but will get filled out by bots that unwittingly try to fill out the entire form.

If filled out, the system knows the form is fake, and will reject the form submission.

Some form providers offer built-in honeypots that you can enable, but here’s a great way of implementing one if your form doesn’t have any advanced honeypot option already built in.

Part 2: Set up a real-time email validator

The second technical implementation is setting up a real-time email validator. These tools check instantaneously that the email being submitted isn’t temporary/disposable, isn’t a known spam email address, is able to block yahoo.com/hotmail.com/etc, and performs dozens of other checks to ensure the email is valid and genuine.

Two of the most popular are Clearout (which has a WordPress plugin) and Zerobounce.

Part 3: Stop serving ads on the Display network (Remarketing is OK)

The Display Network is riddled with click fraud, sometimes perpetrated by what are known as click gangs.

Take a look at where your Display traffic is coming from (within Google Ads on the left side select “Content” > “Where ads showed”), go to those sites, and judge for yourself if you think traffic (or worse yet, conversions) coming from those sites are genuine.

We recommend limited Display traffic to Remarketing campaigns only (i.e. people who have already visited your website), and to make sure you have the first two parts of this section completed before doing so.

Part 4: Stop serving ads with Search Partners

Google Ads offers several “networks” to serve ads in, with the most common being the Search network. But within the Search network, there are actually two very distinct pieces: Google search and Search partners.

Google search is just as it sounds – searches on the google.com (or international equivalents).

Search partners, on the other hand, is far less transparent, and includes searches on sites that “partner with Google to show ads and free product listings”. In fact, Google doesn’t even publish its list of “search partners” – which for a company that preaches transparency feels very sketchy.

This sleuth found a way to isolate domains in the Search partners network (using Builtwith.com), which can be found here: Google Search partners sites

Here’s an example of what a Search partners “search result” looks like on theguardian.com (for “home insurance”):

Search partners presents and issue for two reasons:

• You have no visibility into which partner sites you serve on, essentially wasting money with no way to isolate areas to optimize
• In our findings, the quality of leads from Search partners is VERY suspect. This is especially true if you haven’t take the other steps in this section, as Search partners sites are susceptible to “click gangs” who can, in a similar fashion to the Display network, waste your advertising budget on fake clicks and conversions.

In our opinion, Search partners offers little real, tangible benefit to advertisers, and should be always be excluded/turned off.

Part 5: Stop using Broad match keywords

The final part of this section is probably the most common major mistake we see with Google Ads – serving “broad match” keywords.

Over the years Google has changed the definition of its match types, with the definitions in 2022 being as such:

We recommend ONLY serving keywords with Phrase and Exact match.

When serving on Broad match, you essentially give Google free reign to decide what it considers to be “close enough” to serve your ad. Because you’re only able to see ~25% of Search terms/queries to understand exactly what people are searching for when your ad appears, Broad match presents a bit of a black box, and is a great way to waste money.

One of the other hand, since Google has relaxed its matching criteria for Phrase and Exact match terms over the years, keyword expansion via Search terms analysis is still possible with these match types, even though a seasoned Paid Search professional should be able to cover off on 90%+ of the potential keyword universe when crafting your account structure in the first place.

We can look at the last two parts of this section in a nutshell with this analysis on an account we recently audited:

Within Google ads you can see what people are actually searching when your ad is served/clicked. You can locate this area by going to “Keywords > Search terms” on the left side of the screen within the Google ads interface.

When you hear about finding “Negative” terms, this is where you go to figure out specific keywords/search queries that you’re serving ads on but that aren’t relevant/converting.

However, you can take this a step further by performing an n-gram analysis.

What this means is instead of looking for queries that aren’t converting, look for individual words (within those queries) that while individually might not be large enough to stand out, but when aggregated are slowly wasting your budget. A good analogy for this is “death by a thousand cuts”.

Here’s an example from a recently audited account (key info redacted) . . .

This is a list of search terms that didn’t convert over a 120 period:

If there’s one way to make sure you waste money, it’s by having conversion tracking set-up improperly or inaccurately. This is all too common.

Conversion tracking set-up is sort of like car repair . . . it isn’t rocket science, but is tricky enough that you can easily mess something up if you don’t really know what you’re doing.

There’s no way to tell if conversion tracking is set up properly without a thorough audit, and the best way to do this is with Google Tag Manager’s debugger tool.

First things first – we ALWAYS, if possible, recommend setting up conversion tracking via Google Tag Manager (aka GTM).

If you’re not familiar with GTM, think of it as a shopping cart at the grocery store. When you go shopping, you don’t find one thing you need, bring it to the register, find another, bring it to the register, etc. This would be extremely time consuming, tedious, and if you forgot whether you already grabbed something, you’d have to go back to the register just to double check.

GTM allows you to put everything in your cart at once, easily make sure you have everything you need, and then when ready, head to the register to checkout.

With GTM, you’re able to put all your conversion tracking code (as well as other code not necessarily designed for conversion tracking) in one place, then make sure it’s working properly, and finally publish the code to start being used in the appropriate applications.

We frequently see conversion tracking set up improperly (either doesn’t fire when it’s supposed to, or sometimes fires twice). Here’s the template we use for conducting the conversion tracking audit: