Wanna hear a sad story about some orphans?
Pretty much every company I’ve helped out with paid search marketing has been running tests– in the sense that there were a couple ads per ad group, and they were running on some sort of rotation. But in general, that’s as far as the tests have gone. No one has ever come back to check them. On and on they rotate, through the months and years, speaking their results out into an unhearing universe. So common is this, in fact, that Google introduced a new setting for ads where they rotate evenly for 90 days and then start to optimize (previously, when you picked ads to rotate evenly they would do so indefinitely). They were clearly tired of leaving money on the table because companies weren’t tending to their sad, losing orphan ads.
I know, it’s heart-breaking. Take a moment. Get a tissue.
Marketing Is Not Rotisserie Chicken – You Can’t Set It And Forget It.
It should go without saying, but if you’ve got a test underway you should regularly check the results, promote the winners, and set up new tests. That’s really the only way testing is helpful, in fact.
Even if you’re not testing, you want to make sure that your programs aren’t on autopilot. Running banner ads? Keep your eye on the numbers and make sure they’re still performing. Sending outreach emails? Look at the conversion rates (and of course, test to make them better). Running an affiliate program? Try out different program descriptions and experiment with what kind of offers get the best uptake.
You Think You’ll Remember, But You Won’t.
The other way to make testing helpful? Write down what you’re testing along with the results. It seems fresh to you now, but as the tests and the learnings pile up in your big brain, stuff from a couple of months ago will start to fade away. Plus, writing down the results of your tests means that you will notice when there aren’t actually any results, meaning that you haven’t completed your experiment cycle and gotten to the most important part– learning.
In Lean Startup, Eric Ries describes using a kanban approach to product development; projects are either planned, in progress, built, or validated. Each column is only allowed a certain number of projects, so to move more in some have to be moved out. This system assures that features aren’t being built without being validated.
So it should be with the things you are testing with your marketing. Use a spreadsheet or even a Word doc; write down your what you’re testing, what the results were, and what conclusions you drew from it. If you find yourself with a lot of open tests and no results, go back and close those down.
And in the spirit of keeping your marketing programs alive and fresh– use the baseline performance metrics you got from starting early to critically examine each program you’re running.
Are you finding PPC doesn’t come within shouting distance of your target CPA? DUMP IT.
The traffic you’re getting off of PR placements doesn’t seem to justify your $10,000/month retainer? DUMP IT.
Preparing ads for re-targeting takes you a lot of time, yet the volume it produces is tiny relative to other things you’ve tried? DUMP IT.
The traffic you’re getting from your guest blogging spot is solid and it converts like a dream? DO MORE OF THAT.
And of course, record it all in your learnings document so that six months later the replacement who was hired because you got promoted because you’re so awesome isn’t like, “Why on earth aren’t we doing PPC? So dumb. I’ll start right away!”
Hey, wait a minute… what’s this?
It’s Success Baby! We’re done! Congratulations, you now know everything you need to know to become a flawless marketing stud. People will whisper your name in the halls as you pass, and this time it won’t be because of your horrid yellow socks. Go forth and conquer.
(Photo credit for savory chickens: tomkellyphoto via Flickr. Photo credit for handsome young Clint: some movie studio. Please don’t sue me, movie studio. Photo credit for Success Baby: who knows? That baby is a meme. He’s all over the internet.)