Stop Bleeding Cash: 10 Recent Examples of Poor Optimization

There are two questions I hear which are more related than they appear to be:

1. What does “optimizing an account” actually entail? What are you doing that’s “optimizing” it?
2. These results are a lot better. How did you do that?

The answer to the first one is what gives you the results in the second one. Paid media accounts are like huge machines, and you have to figure out where the cogs are grinding to a halt – what needs oil? What needs a new part? Or does the engineer who built just suck so it needs to be scrapped and rebuilt?

All the factors of an account (bidding, structure, creative, audience, etc) are up for grabs. Trust me when I say, I’m far from perfect. There’s a running statement among PPCers that any account audit will reveal ways to improve, and it’s true.

But sometimes…they’re just too obvious to ignore or make an excuse for.

When you work agency side, you see all kinds of crazy stuff when you take over a new account. It’s usually brought by a client who has a gut feeling things could be going better, but they don’t know enough to be able to figure out why their gut is telling them that. So the new agency digs – sometimes you don’t have to go very far.

I should point out there are some reasons this can happen that might not be obvious when you look at the data. Those include:

1. Clients can make really weird mandates. Keywords they insist be bid on, creative they demand continue to run, etc. That kind of background color isn’t obvious when you are dealing in cold, hard data.
2. You don’t know the agency/client relationship. Maybe the agency took on a relatively small client for them and hours and service are limited, so they focus their efforts in certain ways. (I still blame the agency or PPC manager here – stop taking on clients that are a bad fit for you.)
3. Lack of expertise. An agency puts the new kid on it, or it’s a fledgling PPC manager’s first true account b/c their buddy started the company. Who knows, but inexperience is a results-killer.
4. Forest for the trees. PPC managers can be detail-oriented to the point where they miss extraordinarily large things. It’s like they are focusing on getting a loose doorknob on one cabinet door to be perfectly tight while termites are eating the rest of the house.

On Twitter a couple weeks ago I jokingly that I should create a blog post called “Stop Bleeding Cash” and it would just show screenshots of accounts losing money over something that should seem obvious. People said they’d be interested in seeing it, so here we are.

I deliberately chose instances that didn’t have weird circumstances, with data found in a simple report…nothing fancy, no crazy-scientist analysis. Just normal data that any capable PPC manager would be looking at on the regular.

Also, I have multiple examples of the same thing – I did this purposefully so folks understand these aren’t weird, isolated incidents. We see this stuff way more than you’d believe.

Without further ado:

 

Devices? What Devices?

Who needs modifiers when your CPAs are drastically different? Not this account, apparently.

….Or This One.

This still amazes me. Everyone talks about device behavior, so how is nothing done with it?

And Then, When Bid Modifiers Are Just Wrong

Repeat after me: negative bid modifiers for poor performers. Positive ones for good performers.

When In Doubt, Do….Nothing?

Clearly, the same bid is a great idea, here. Position 1 is obviously doing wonders for the bottom line.

Females Are Having None of Your Sh*t

…And They Aren’t Here, Either.

The Stalker Method

“YOU WILL BUY MY PRODUCT. BUY IT!”

When you have AdWords Optimizing Ad Rotation for Clicks

Good thing we aren’t showing our converting ad the most often. Whew!

Speaking of ads…

Psst….pause the underperforming ones. You’re welcome.

Oh hey, we still don’t understand bids over here…

What is this “goal CPA” of which you speak?

This Is Normal And Shouldn’t Be

THIS is what you look at what you optimize.

THIS is what makes your results as a PPC manager stand out, because the worst part about all this is: it’s more normal than not. Accounts are more likely to be riddled with  junk like this that would get fixed with an hour or two of taking a step back and pulling the right data.

It’s also why I love what I do every day…the opportunity to fix this, stop the bleeding, and get clients on a better path to profitability is something that hasn’t gotten old in over 10 years of doing it.