To Test Or Not To Test

The moment of truth… A first month media plan for a big-name (or soon to be big-name) client.

They are looking for big. They are looking for bad. They are looking for kick-ass creative opportunities with all the bells and whistles.

The client is looking for the most targeted placements. They want lots of visitors. They want branding. And they want you to achieve cost-per-acquisition goals that would make the producers of Ginsu knives sweat bullets.

And they have a small budget!

How do you deal with a client that has got a Chez Pennise palate but only a McDonald’s wallet? What do you show a client who can’t stop thinking about branding, but imposes upon you aggressive direct response goals?

Well, first things first.

One of the biggest tasks of an online media planner or buyer in the industry today is managing client expectations.

Pressure is on from a Board of Directors and VCs (usually one and the same) to grow traffic and move units. Everyone thinks they are going to be the next Amazon-of-insert here, and it can be done with a limited budget.

And finally, the client wants to be a recognizable brand in the most traditional sense, but they are judging your success by a cost-per-action metric.

The reality is that you need to get the client to set linear priorities.

Branding and awareness or direct response objectives have to be agreed upon first. And only one of these things can be allowed to serve as master.

When pressed, nine times out of ten, a client will always come to realize that the buy you put together for them needs to satisfy a direct response strategy because their success (read YOUR success) will be judged by a cost-per-action metric.

So the client has agreed; the first month’s buy will be dictated by a direct response strategy. Maybe against their will, since direct response has long had that unsexy quality of junk mail, Select Comfort ads, and K-Tel’s AM Gold.

But let’s face it, the accountability afforded by the web lends itself so perfectly to committing direct response campaigns.


Yet the client still wants you to be aware that they are trying to build a brand and that needs to begin being achieved in tandem with yielding efficient cost-per-acquisition.

How to proceed?

To test or not to test, that is what the question is now. And the answer out of any good planner’s or buyer’s mouth should be, “Test.”

Given the vast array of web sites and a significant amount of them with inventory available at low CPMs, there is no reason why you can’t take some of these sites for a test drive.

Find some categories that you suspect will do well against your target and choose one or two sites from a long list from each of those categories.

You can make your category and site selection using intuition, experience, or the use of some syndicated proprietary research such as @plan.

For the first month out, pick the sites that have the audience you are looking for and come at a low CPM.

I understand cost isn’t always the reason to make a choice for or against a site, but you really want to minimize your exposure first time out of the gate. If sites are performing well, bully for you! If not, well, at least it didn’t cost you a small fortune in branding opportunity sponsorships to find out.

After running on the sites from your selection, you can then read data that will start to point out the sweet spots for your client. I like to call them “islands of performance.”

Follow up this first buy with a second that includes your winners (whereby you can confirm that the first month wasn’t just a fluke), takes the reserve from your losers, and invests it in some new sites vertical to those that performed well.

And if you find yourself with a bit left over, go for something a little more “out-of-the-box” on one of the clear performers. This allows for the continued effort of accomplishing numerical goals efficiently while at the same time being smart about where you place “branding” dollars.

A test strategy gives you and your client the opportunity to figure out what sites yield good results according to a direct response objective before upping the ante with higher out-of-pocket sponsorships or other branding opportunities.

By the end of your third month of buying, the sites can then be fulfilling both a direct response and branding objective with the same dollar.

Related reading