Are search keywords artificially high or artificially low ?
Every six months, for the last year and a half, Jupiter Research has run an executive survey of search engine marketing (SEM) advertisers and agencies. The current edition of the survey, now underway, is our biggest ever.
Of all the questions we ask, my favorite is "What problems do you encounter with search engine marketing?" It’s perhaps the most useful, especially to agencies specifically hired to solve problems and relieve pain. Knowing what frustrations advertisers experience is invaluable, especially when walking into a pitch. For the last few rounds, the most popular answer to this question has been: "Desired keywords are too expensive."
On the face of it, this is a difficult problem to solve. If you’re a small advertiser and want to bid on "dvd player," chances are you can’t afford it. The marketplace for keywords is supply-and-demand driven, and demand for certain high-traffic, high-converting keywords is huge, driving the cost up to a level at which only the biggest players can compete.
This is nothing new for advertisers. I’m sure a lot of small companies would like to place an ad on Time Magazine’s inside cover, on the corner of Broadway and 42nd St., and during "American Idol." But they can’t afford it. Valuable inventory commands a premium.
Yet real buzz is growing that keyword costs are rising across the board. Publicly disclosed Yahoo financial information bears this out. Average revenue per keyword increased over the last few years, and Jupiter predicts the trend will continue at least until 2009. Does an increase in keyword costs necessarily translate into a decrease in keyword return on investment (ROI)? It better not. If it does, you’ve figured your ROI structure wrong.
Demand Will Increase
I was chatting recently with the founder of a big SEM agency. He tells me that, even today, he goes into meetings with marketing departments at Fortune 50 companies and explains how Google makes money. "People pay for those listings on the right." Really?! Good for him. He then explains how to get into the game.
Anecdotal evidence, but it definitely tells us the demand for keywords is artificially low, as the set of bidders is incomplete. Imagine a huge bike race without Lance Armstrong competing. The race’s winner would know he wasn’t challenged to the absolute theoretical maximum, though I’m sure he’ll accept, and no doubt celebrate, his win.
That’s precisely what’s happening with keyword bidding right now, and why I worry advertisers perceive artificially low costs as a stable market and build ROI calculations from them. If you’re paying less than you should for a keyword, that difference isn’t real ROI. It’s bonus. It’s gravy. But if you base your rationale for using SEM on it, you’re in trouble. That difference is in danger.
Needed: Bottom-Up ROI Calculation
Here’s what you should do (this is straight-up direct marketing): Before you bid, or even choose advertising vehicles, you need to know how much you can spend on a lead. This is a bottom-up approach, because you must start with the cost of the product you’re marketing; figure the amount of overhead that exists in producing, packaging; and delivering that product; add in the profit margin that makes it all worthwhile; then identify what you can comfortably spend on marketing.
Let’s say you’re selling a clock for $10. You want a 20 percent margin, so you’re immediately down to $8. You’re buying the clock wholesale from the Black Forest for $7. That leaves you with $1 to comfortably spend getting leads. You may now go out and search for any opportunity that will deliver leads for $1, including search.
Unfortunately, too many marketers haven’t taken these steps. They determine bid prices top down, which usually means going to the engines, inputting a keyword, and seeing how much it costs. Most supply chains are quite a bit more complex than the clock example, so you can see why the task hasn’t been undertaken. Not only is it way too difficult, it’s also outside the responsibility (and skill set) of the marketing department. Generally, they’re in charge of advertising. So we get rough guesses instead of bids.
The clock seller, though, is well prepared to develop a bidding strategy. If he learns bids for "clock" run $1.25, he knows that’s not worth it for him. Maybe he’ll bid instead on "black forest clock." If that phrase is only $0.50, he’s in great shape. He makes his margin, but also gets a half-buck bonus on each sale. If "black forest clock" doubles to $1, too bad. But he’s still totally comfortable, as his ROI is built bottom up and untouchable.
The clock seller’s in good shape until "black forest clock" goes over $1. He knows this is an untenable situation, and he’ll have to do one of a few things: eat the cost (i.e., take it out of his margin); shift to another keyword; make adjustments in his costs; or be a better clock seller.
The first option should immediately be off the table. Don’t eat costs, at least not at first. The second is a good idea, but your strategy shouldn’t be to consistently move about trying to find good deals. The third and fourth are the best bets and represent the only real, long-term strategic position. Essentially, they’re the same: focus on efficiency.
As a marketer, you can’t control the cost of keywords, but you can control your own business. The more efficient you are, the better bidder you can be. Imagine the clock guy found a smaller shipping company to work with and was able to decrease his per-unit import cost by $0.50. He could take that $0.50, increase his bid and be more competitive because he’s more efficient -- and he hasn’t touched the bottom line.
Similarly, and perhaps most important, he can improve his site to make sure it’s converting at its highest level. Built into his overhead is the pure cost of making a sale. If he’s an online retailer, this is all held up in server costs. The prime metric of server costs must be the number of sales made per day. The more sales, the more efficient he is, and (again) the more money can be put toward bids and the more competitive he can be.
Tools Are Necessary!
As keyword costs increase, inefficiencies will be laid bare. They’ll be presented to your competitors as weaknesses. Get a handle on them before you begin to see keyword costs rise. As you identify them, you can put together the cost-per-lead you can comfortably pay and optimize your entire campaign around that.
When you’re ready for that, you’ll realize a select few organizations offer tools you can use to actually put this strategy into place and make it work. Most recently, iProspect released iProspect Search Engine Bidding Agent (iSEBA), an agent-based tool that takes the work you’ve done to identify your target cost per lead and organizes and optimizes your entire campaign around it.
The next phase of SEM will be about efficiency in your business. It will be galvanized by marketers who experience an increase in per-keyword costs. Now’s the time to get ahead of that wave.
Revolutionize your digital marketing campaigns at ClickZ Live San Francisco (August 10-12)!
Educating marketers for over 15 years, our action-packed, educationally-focused agenda offers 9 tracks to cover every aspect of digital marketing. Join over 500 digital marketers and expert speakers from leading brands. Register today!
Gary Stein is SVP, strategy and planning in iCrossing's San Francisco office. He has been working in marketing for more than a decade. Gary lives in San Francisco with his family. Follow him on Twitter: @garyst3in. The opinions expressed in Gary's columns are his alone.
US Consumer Device Preference Report
Traditionally desktops have shown to convert better than mobile devices however, 2015 might be a tipping point for mobile conversions! Download this report to find why mobile users are more important then ever.
E-Commerce Customer Lifecycle
Have you ever wondered what factors influence online spending or why shoppers abandon their cart? This data-rich infogram offers actionable insight into creating a more seamless online shopping experience across the multiple devices consumers are using.