Commentary

Yes Sir, Can I Have Another: How to Avoid A Search Spanking

We were in the airport, on our way back from a client meeting. Our conversation veered towards formative experiences from grade school. One colleague regaled us with tales of corporal punishment from his years at a strict religious school. "In shop, we had to make our own wooden paddles. We were whacked with our own paddles when we broke the rules. We even had to drill holes in them, to increase the force with which they struck."

What irony! Imagine--tasked with creating the weapon of your own demise. Damned if you do, damned if you don't: create a good paddle, get a good grade, and a more painful whack. After all, everyone gets paddled sooner or later. I surely would have been.

My schadenfreude evaporated as I realized that in search engine marketing, many of us tend to create our paddles every day. And we drill holes in them to make them smack especially hard. Let me assure you that yes, my mind actually did make this connection. If I tell you that I was coming from a client search marketing meeting, on a few hours of sleep, you may be more apt to believe me.

advertisement

advertisement

We create our paddles in the form of the massive amounts of data on which we frequently report, and all the tactical maneuvers in which we become engaged. Having spoken to many search marketers on the client side, I am pretty confident that these paddles do not exist solely in agencies, but in any organization where individuals are tasked with creating and executing paid search engine marketing campaigns.

How do we build these self-punishing paddles?

"You bet I can tell you click, cost, and conversions for every keyword."

"Sure, I can look into why you were in third position instead of fourth for the keyword 'pink spectacles' in Google."

"Report on keywords by category, by search engine, and by click volume? No sweat."

We get whacked by letting these tactical, laborious processes take the place of what drives real performance. After all, none of the tactical deliverables above actually help drive performance. While data reporting and business intelligence is a fundamental part of any search program, gratuitous grunt work can become a parasite that invades real performance.

What is search performance? Certainly the list of the metrics and their nuances differs by marketer. Still, I would submit that almost any major advertiser in paid search is driven by two metrics above all else:


1) Volume of orders/sales/inquiries/impressions/dialogues/etc. Call this volume for short.
2) Marginal cost. This is how much each order/sale/inquiry/impression/dialogue costs. Some calculate CPA (cost-per-acquisition) or ROAS (return-on-ad-spend). Let's call this marginal Cost.

I do want to emphasize that there are other important factors in any search campaign, such as how business intelligence from search can inform other media strategies, or brand impact. However, most advertisers expect, above all else, a maximized volume and a minimized marginal cost.

What drives volume and marginal cost? Well, it's not a masterful 8th-dimesional pivot table. It's not the dogged pursuit of why the keyword "pink spectacles" on Yahoo is mysteriously absent at 9:41 p.m. The drivers of performance in search are really four-fold:


1) How you bid
2) What keywords you have
3) Your creatives and landing pages
4) The search engines with which you advertise

Sure there are more drivers we can think of, but generating reports on every keyword, troubleshooting the whereabouts of that elusive "pink spectacles" keyword, would not likely appear in the top 50.

Not only do these non-material activities keep us from our primary focus, but their execution only serves to worsen the situation. One new keyword-level report structure turns into a "great idea" to generate three additional, similar reports. Before we know it, we can spend more time reporting than on creating effective strategies.

So why do we do it? Why do search managers spend hours--sometimes days--creating these lengthy reports, or spend countless hours investigating what caused a minor keyword to behave inappropriately?

From my vantage point, we tend to craft these "paddles" when making one--or more--of the following mistakes:

First, we can't wait to share data. Or, we make the mistake of equating data with performance, and we train our clients to do the same. As any search expert will admit, we live and die by data. But just as a homebuilder does not provide a daily inventory of the type and quantity of nails used, nor should a savvy search marketer spend valuable time taking inventory when she could be analyzing, optimizing, building. But for some reason, we can't wait to share this data.

Second mistake: we do not always recognize that a search program has become a time-suck because we have lost (or never gained) the confidence of clients, internal or external. They need to poke around into the details to get comfortable. In this case, we should realize that the request for granular data or tactical hoop-jumping is not so much a problem as it is a symptom. Fix the confidence issue by making sure you have the right resources focused on the right job, and that you're effectively communicating performance to the folks in charge.

Third mistake: sometimes, we just don't know any better. We train ourselves to say "yes" to clients, "yes" to our colleagues, "yes" to our bosses. While saying "no" can get you into trouble as well, it seems that search marketers would be better off helping their clients (or colleagues, or bosses) to ask the right questions. The ones to which we can say "yes." And not, "yes, sir, may I have another."

Sure we need to report. We need to report on performance (usually, volume and marginal cost.) We need to share the key learnings and business intelligence metrics that we gather as search marketers. And yes, we sometimes need to spend time researching a particular keyword, or responding to an ad-hoc request. But regardless of whether you work at an SEM, at an interactive agency, or on the client side, your resources are limited. To allow those resources to shift away from the key performance drivers and towards reactive reporting or endless ad-hoc requests will hurt the performance of your search campaign. And you will need to prepare to get spanked by all those masterful weekly 150,000-row reports you cranked out.

Next story loading loading..