Paid search managers serve many different roles these days, and that can be harmful to the program if staffing isn’t adequate to handle the extra demands.
This is a product of paid search’s success. It has become hugely important as a driver of customer acquisition and wallet share, and as such has drawn the attention of folks from other parts of organizations who don’t always understand how paid search works. These folks often fall victim to commonly held misconceptions.
Six common fallacies about paid search campaigns
- More is better. Many folks assume that if a program is doing well, then throwing more people at it will make it better. Often, that has the opposite effect. Just as too many chefs in the kitchen creates problems, too many people trying to fill their time by “doing stuff” can create problems rather than solving them. Priorities get lost, and the people who know what they’re doing end up doing less of what works and spending more time answering questions, preparing reports and fixing misconceptions.
- What is most visible and easiest to understand is also most important. Good paid search managers are often pushed away from the highest leverage activities because they are complex and nuanced, and other folks in the building don’t understand them. They are often forced instead to continually come up with new copy tests and restructure accounts looking for a magic bullet, even when quality score signals suggest there isn’t much opportunity. The folks who are farther away from the day-to-day can see the ad copy, but they don’t take the time to understand the importance of match types, syndication networks, anticipatory bidding, etc.
- The more granular the view, the better. The people in our shop are famous for being data hounds, but part of being a good scientist is knowing when to use the naked eye, when to us a magnifying glass and when to use a microscope. Studying a TV with a magnifying glass doesn’t improve the picture; rather, it turns it into meaningless red, green and blue dots. Trying to glean insights from keyword level data from a single day is usually a similar exercise. Reacting to statistical noise leads to yanking the proverbial rudder back and forth which has the net effect of slowing the boat.
- Studying data = analysis. There is an important distinction between studying actionable data and studying trends over which the paid search analyst has no control. Studying clickthrough rate trends and cost-per-click trends are examples of the later. Since the impression counts vacillate with the whims of the engine’s match types and network partnerships, CTR tells us nothing that leads to intelligent action. Quality Score tells us something about the quality of the copy, but even that info is blended with information about the commercial intent of searchers who see that ad. CPC variances likewise are a function of market fluctuations outside of the advertiser’s control. Advertisers control their own bids, but not the bids and quality scores of those below them on the page—the ultimate determiners of cost per click for a particular keyword. There is a natural human tendency to try to explain what just happened at the expense of focusing on what is within one’s control.
- Competitive analysis helps inform paid search decisions. It’s certainly worthwhile to poke around the programs of competitors at the top of the page. Are their prices or product selection better than yours? Does that give them higher conversions and allow them to spend more to attract traffic? Are they touting promotions in their ad copy that you could also tout, but don’t? These are good questions to answer, but beyond these you can drive yourself nuts trying to gain actionable insights from a competitor’s program. Ultimately, there is no way to know whether the folks at the top of the page for any given keyword are marketing aggressively and profitably, or aggressively and stupidly. The normal assumption of top executives that the other guys are geniuses and their own marketing team is full of morons may produce a healthy drive for success, but often results in huge wastes of time.
- Strategic plans and progress reports help drive performance. As I wrote last year, paid search is more about process and methodology than it is about planning. A great analyst needs to have the freedom to follow his or her nose to find the truffles, and just as a stock analyst shouldn’t be asked to determine what stock to buy 90 days from now, a paid search analyst will have a hard time telling you what the highest leverage activity will be three months down the road. Methodical practices for testing, keyword maintenance, seasonality research, etc need to be balanced with time for deep data dives. Well prioritized methods produce better results than any grand strategic plan.
It may seem that pulling extra reports together does no harm, but it certainly can.
It’s kind of like an old dandruff shampoo commercial from the early 80s:
The announcer says: “Our shampoo has the same medication for treating dandruff as our competitor…” [two beakers are filled to the same level with ‘medicine’] “…plus an extra ingredient that makes your head tingle!” [A shot of head tingle goop is added to the advertised brand’s beaker.]
I always wanted to scream: “But the customer doesn’t use any more shampoo, they use the same amount! The additive isn’t ‘extra,’ it dilutes the stuff that actually treats dandruff. You don’t get more, you get less!”
Anytime resources are constrained—and resources are almost always constrained—misusing those resources means incurring opportunity costs. Now, it may very well be the case that appeasing all the internal constituencies interested in (if not knowledgeable about) paid search is worth the cost of hiring enough bandwidth to cover the basics and the extras as well, but if staffing isn’t adequate, be careful not to make room for desert by jettisoning the main course.
Next time, I’ll give a long list of menu items that make up a healthy main course.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.