If you’re not careful, your PPC budget can disappear at the blink of an eye. Here at WordStream HQ, we’ve inherited hundreds of accounts that have fallen into these perilous, money-wasting traps. Here are the five craziest real-world scenarios our account management team has encountered, with tips on how avoid them in your accounts!
About two years ago, we brought on a client who had two vastly similar campaigns. One was labeled “Bath Products” and the other, almost identical campaign was labeled “RLSA—Bath Products.” The RLSA campaign was spending nearly as much as the standard campaign and really wasn’t yielding better performance—both of which raised red flags.
Upon further investigation, we discovered that the account manager who originally set up that so-called RLSA campaign had actually just cloned the original campaign and added the audience as a “bid only” audience, rather than a “target and bid” audience, so it was still serving ads to anyone who matched to the keywords, regardless of whether they’d been to the site before in the past.
Despite its label, it really wasn’t an RLSA campaign at all, it was just a duplicate campaign that was effectively bidding against the first. In this particular account, after running the faux-RLSA campaign for 30 days and spending $1,700, only $1.73 had been used as intended, to reengage with a returning audience.
Of course, this was a super-simple fix. We adjusted the settings so the campaign’s ads were only served to those on the remarketing list.
This is an easy mistake to make, but it can certainly wreak havoc on your account. Before using RLSAs, be sure to understand these two, vastly different options. By selecting target and bid, you’ll only show ads to people on your remarketing lists. With bid only, you’re essentially adding the remarketing list in along with your standard search targeting, but have the opportunity to set specific bids for the remarketing audience.
Years ago, a client that I was consulting for was mysteriously hemorrhaging money. His account was performing perfectly and then, at the drop of a hat, things took a turn for the worse. All of the sudden, impressions were at all an all-time high, click-through rates had plummeted (although overall click count was slightly elevated) and conversion rates were down. It was a code red situation.
The two of us wracked our brains trying to figure out what the heck had gone wrong. We were at a loss—we’d made no major adjustments to the account, no competitors had “dark-horsed” their way into the auction and nothing notable had changed in his industry.
As we frantically tore through his search query reports looking for clues, we stumbled on something shocking—his ads had been showing for a slew of Justin Bieber-related search queries. The client’s company sold fiber optic cables, so this definitely did not add up.
As we did a little more digging, we unearthed the full story. Nearly all of his keywords included the term “fiber.” Through close variant matching, Google deemed “Bieber” to be a misspelling of “fiber.” Things really heated up when Justin released a new album and Beliebers were pounding the SERPs in search of it.
Bieber fever is real, guys. Here’s proof.
Not surprisingly, these pesky Bieber fans were definitely not in my client’s target market, so we set “Bieber” as a negative keyword—costly crisis averted!
This client’s account was set up perfectly and yet, he still fell into this sticky, costly trap. When it comes to close variants, it’s nearly impossible to predict what types of terms Google will match you to. More often than not, you’re linked to relevant queries that are truly misspellings or close variations of your targeted terms…but that doesn’t mean that a few bad apples don’t make their way through, too.
The more negative keywords you can predict, the better. For those poor matches you don’t foresee, the best defense is to review your search query reports on a regular basis. The sooner you catch them, the sooner you set them as negatives.
Your targeting settings matter—in a big way. Far too many advertisers neglect to set these up from the get-go, resulting in tons of impressions and clicks from unqualified regions. One of our consultants, Francine Rodriguez, recently worked with a client whose targeting settings cost him thousands of dollars in wasted ad spend.
This particular client sells fan paraphernalia and jerseys for various soccer teams online. Given that there’s a global audience for these products, he set his targeting settings to “world.” Not surprisingly, he exhausted his budget rapidly—as it goes, there’s a fair number of people around the world searching soccer-related terms.
His traffic yielded its fair share of impressions and clicks, but little to no conversions. As he and Francine reviewed his geo-reports, they found that the vast majority of these impressions/clicks were coming from India and China. The problem is, the client’s product pages were in English and his pricing information was only listed in USD, CAD and Euros. Even if these visitors did consider making a purchase, the shipping costs would be so exorbitant that they’d likely outprice the order itself.
Francine helped him narrow his targeting to areas that made more sense for his company, which drastically improved his performance.
Just because you can sell to the whole world doesn’t mean you should. Understand the regions that are most likely to follow through with a purchase and focus on those first. Over time, you can always expand your reach to new areas (cautiously!).
Be sure to take advantage of Google’s sophisticated targeting options and remember, when it comes to geo-targeting, Google will automatically default your location settings to focus on people located in or searching for your terms in your country (if you’re in the US, it includes both US and Canada in this default).
We PPCers LOVE running experiments. We test ad and landing page variations, audience combinations and campaign types fanatically. The only thing that can get in the way of our endless experiments is our clients. The vast majority of clients push us to stick with what works, rather than going the riskier route and trying something new.
So, when Navah Hopkins, one of our PPC consultants, learned that her new client was all-in on running tests in his account, she was thrilled. Together, they dreamed up all sorts of crazy campaigns to test for his account. They were risky, but if they worked, it’d be a huge breakthrough for him. She instructed him on how to implement the changes and set up a follow-up call to assess the results a few weeks later.
When Navah jumped in to check-up on the progress of the tests, she was horrified to see that the account had blown through tons of money. Upon further investigation she discovered the fatal flaw—her client had missed one small step of the setup process. He’d allocated the same budgets for his test campaigns as his core campaigns.
Luckily, a few of the tests had worked well, mitigating the impact of those that didn’t. But, this scenario could have been disastrous.
Running tests is the best way to determine what really works in your PPC account. However, you should only allocate a small percentage of your overall budget to experimental campaigns, to protect yourself, should they backfire. (More tips on how to allocate your marketing budget here.)
Is Kim Kardashian singing the praises of your luxury handbags? Is Lebron recommending that all aspiring athletes invest your flashy sneakers? Congrats. You’ve scored yourself a celebrity endorsement, something many brands would kill for.
In 2003, McDonald’s paid Justin Timberlake $6 million for an endorsement deal, which included shooting a music video and recording a sound clip for the “I’m lovin’ it” jingle.
One of our senior SEM managers, Jaclyn Jordan, works with a beauty product vendor who has the backing of an uber-popular celebrity (I’ve been sworn to secrecy, so I can’t reveal the name). Her endorsement has had a ridiculously powerful impact on the brand and we reap it for all its worth. We’ve adorned all of their display ads with her beautiful face, the text ads remind readers that she’s a big fan of the brand and we even bid on keywords that include her name.
Between this American sweetheart’s endorsement and Jackie’s brilliant PPC skills, this account had been killing it for months. Each week’s CPA was lower than the previous until—BOOM—the celebrity endorser got a haircut. The second the news broke (yes, I realize how absurd this statement is, but it was covered by multiple news outlets!), the account’s CPAs skyrocketed. Jackie, not privy to THE NEWS was shocked to see performance plummet. It wasn’t until she dove into the search query report that she realized the source of all of these new, unqualified searches.
In a perfect world, your brand’s token celebrity would give you a heads up anytime they do something that might cause a spike in searches for their name. But, chances are, they’re not going to turn over their personal calendars to protect your PPC account.
The best defense against scenarios like this is to monitor your celebrity closely. Try to prep for major events (awards shows, album releases, big games) in advance. For unexpected news (like haircuts), set up a Google Alert for their name, so you will be notified any time something newsworthy occurs. Once you know what you’re up against, you can set negatives or tweak match types to limit unqualified traffic.
Please read our Comment Policy before commenting.