Taking Work and Waste Out of B-to-B Search Marketing

Dec 14, 2006 7:34 PM  By

(Searchline) The great basic advantage of search marketing directed at consumers is that users self-qualify: If they put into that query box the keyword that brings up your ad or the organic link to your page, then chances are strong that they’re interested in what you have to offer and may in fact be ready to buy.

Ironically, that can be what makes business-to-business search marketing so difficult. Because the process of buying for businesses is so different from that on the consumer side, advertisers looking to sell to enterprises can misstep in designing their ad copy, their landing pages, or their offers.

B-to-B product searchers are often in a mode of “considered purchase,” according to Karen Breen Vogel, CEO of ClearGauge, a Chicago-based online marketing firm that specializes in b-to-b campaigns for both Fortune 1000 companies and small and midsize businesses. In this environment, searchers are not likely to end up going to your Website and buying something from you immediately.

“They are more likely to being going through a process of investigation and research,” she told a session at the Search Engine Strategies conference in Chicago earlier this month. “And they probably have a lot of other people at their company involved in that consideration process.” In fact, she pointed out, marketers in the b-to-c world have lately come to realize that they too were often marketing to people who did not buy immediately and had to recast their own search marketing approach to take account of drawn-out buying cycles.

One of the first challenges for b-to-b marketers is simply figuring out who is searching on their keywords, what roles those searchers play at their companies, and what their intentions are. “There are just so many people out there in the buying cycle–influencers, decision-makers, financial buyers, technical buyers, end users–and people searching on the same word can have totally different intents in terms of what they’re looking to do and what their expectations are,” Vogel said.

The answer, she told the audience, is to give people who come to your site through search a number of ways to self-segment. She showed as illustration a landing page for a Dow Chemical division that manufactures foam for products. The left side of the page contains a list of ways visitors can define themselves in terms of how they use that foam, from molded-foam producers to bedding-foam producers. It’s a way for them to tell Dow who they are and to be directed to a page that will be more relevant to their needs.

The Dow foam page also included both a “contact us” e-mail button and an “ask a question” button. Most people understand that “contact us” usually results in a sales call, Vogel said, and therefore they don’t use it if they’re not ready to buy. But letting visitors ask a technical question is another way to segment out the early-stage researchers and visitors who may be having a problem with their current foam products and seeking solutions on the Internet.

Hindsight is another way to segment site visitors in b-to-b. Vogel pointed to a Siemens landing page that, it turned out, was attracting particular traffic from technical managers at factories. ClearGauge built the page, set it running, and then watched to see which keywords were bringing visitors to that page, rather than to one that would appeal to financial managers, for example.

Another hurdle in optimizing b-to-b search marketing is the relatively low sample sizes available for testing. B-to-b sites usually don’t attract anything like the traffic that consumer-facing ones do, and that can make A/B or multivariate testing yield unreliable or even misleading results.

Vogel recommended simplifying tests, perhaps by starting with A/B rather than multivariate testing, and avoiding tests on pages aimed at broad audiences; that should mean that a small sample size can still produce statistically meaningful results. “We’ve seen people doing multivariate testing who’ll have to wait a year or two to get the volume to produce meaningful results,” she said.

For b-to-b marketers who want to test creative, offers, or pages, Vogel recommended using the Google AdWords validity rate checker made available online by testing firm Vertster. That tool uses relevance theory to help determine if you’ve amassed enough campaign history to determine with confidence that test version A is performing better than version B.

If consumer search is plagued with click fraud, Vogel said, b-to-b search marketing is afflicted with an epidemic of click waste: search ad clickthroughs by visitors who are not qualified and merely burn through advertisers’ pay-per-click budgets without purpose. “A lot of marketers are on the big search engines and are not filtering out clicks that will never buy,” Vogel said.

One remedy for this click waste is to make sure that the appropriate vertical search engines are part of a long-term b-to-b SEM program. “Clicks from a Google or Yahoo! that don’t convert will probably never convert,” Vogel said. “Unconverted clicks from a vertical search engine in your field are more likely to be people you want to do business with but who just didn’t happen to take that action today.”

Profinet, an enterprise communications platform from Siemens AG, compared metrics of six-month-long marketing campaigns on both the Google general engine and on Business.com, a very broad but still vertical business-oriented search engine. Over that time, the Business.com effort showed 60% more conversions than the Google campaign at one-fifth the acquisition cost. It also produced 14% more page views than Google–another key metric in determining b-to-b search effectiveness.

Vogel also showed a search ad for GE’s commercial finance division that specified in the copy that it had offers for prospects looking to borrow more than $10 million. Despite that wording, GE had a lot of traffic coming through for people looking for as little as $500,000. So the ad’s landing page contained another filter that asked click-through visitors how much they were looking to borrow. If they said less than the target loan, they went directed to a send-off page that pointed them to a GE division better suited to handle their needs. GE and ClearGauge also tracked the keywords that brought those wrong visits to the site, with the intention of cutting those terms off the keyword list.

Finally, with less visibility into outcomes than their b-to-c kin, b-to-b marketers are harder pressed to calculate the true cost of producing a desired action. Vogel recommends taking stock of all the “highly valued activities” a visitor can perform on a b-to-b site that give an indication of growing interest in a product or service.

To do this, you need to build your Website to operate like a face-to-face consultative sales call. At the end of the encounter, the prospect should have moved through some of the same milestones that matter in a sales meeting, Vogel said: “They were able to tell you their business problems and to get a little value, and you were able to get something from them, and to get their ‘business card’ at the end by getting some information. All these things can tell you that something better is happening than just someone landing and leaving.”

The milestones will vary from one industry and one company to another, but each b-to-b marketer should figure out what they are and then assign point values to show their relative importance in the buying cycle. Landing on a page could equate to awareness and be worth 10 points, while filling out a specific form could mean active interest and be worth 20 points.

Assigning those values, and then tagging not just Web pages but also form fields, will let you use their Web analytics tools to measure more than clicks and sales. At the end of a day, you’ll be able to add up the total value of the actions taken on the site and have some idea of the value they created. (If you’re not into assigning points, Vogel conceded, you can settle for a “win, place, and show” tally, with conversions of any kind as wins, page views as place signifying engagement, and simple visits as the show position.)

“The important thing is to put relative value on behaviors and not just say that everyone who landed on your site is equal,” Vogel said. By setting up a hierarchy of site behaviors, you can get around the problem of not seeing to the sale outcome; you can calculate cost per point and determine which campaigns and keywords are bringing in the best returns, not just the greatest number of clicks.

The work involved in valuing visitor actions and setting up point systems may pay off sooner rather than later, Vogel pointed out, because the big general search engines are showing signs of adding a cost-per-action (CPA) model to the current cost-per-click one, both to cut back on fraud and because they feel it will appeal to marketers in the long run.

When that happens, she told listeners, “You will have difficulty knowing what you want to bid in a CPA model if you don’t know what that action is financially worth to your company.”