Regardless of who you are or what you’re looking for, one of the most common ways to look for products and services is the mighty internet. An unpublished BioBM study found that among life science researchers, 45% will turn to search engines first when looking for a product or service – roughly the same amount as will ask a colleague first – and almost all scientists will perform an internet search at some point in their buying journey. Given the near-ubiquitous prevalence of search as a tool to find products and services, search engine marketing just seems to make logical sense. If you have a product, and someone is looking for that product, then put up an ad, they’ll click on it, and bingo – for a few bucks you’ve targeted a highly relevant member of your target market who is even looking for product information right now! Simple, right? Not always…

There are, in fact, multiple scenarios in which search engine marketing can fail. One of those reasons, however, is a bit more difficult to detect and can actually cost you a lot of money.

eBay Research Labs recently published a study where they set out to determine if brand keyword search ads, in other words keywords that contain the brand name of the company, were worthwhile. Unsurprisingly, they found that such advertising was not effective; in these circumstances people were using Google as a navigational tool and when paid search was turned off, and therefore paid traffic dropped to zero, their organic traffic increased by roughly the same amount.

The much more interesting question that they asked was: “What would happen if we simply turned search advertising off altogether?” The answer to this may seem obvious. If someone searches for “used Gibson Les Paul” (an example they use in the paper) a number of guitar resellers appear in organic search prior to eBay. As this is also the case for many other product-specific terms, eBay’s search ads help direct traffic to eBay when they would otherwise be directed to other sellers / resellers, and thereby increasing eBay’s business. It seems to make logical sense.

eBay wasn’t satisfied with that assumption, however, so they took a sampling of United States geographies and turned off all search ads, leaving search ads in the rest of the country on as a control. What happened to their sales? Largely, nothing. Looking at the sales and advertising data in conjunction with customer data, they found that search advertising is only cost-effective on the least active customers; those whose last eBay purchase was not recent and who made few purchases in the past year. However, eBay is a very popular company and those infrequent purchasers constituted a small percentage of searchers. Therefore, when cost effectiveness was calculated, search advertising had an astonishing -75% ROI. In other words, for every dollar they spent in search advertising, they got back only 25 cents!

Most life science companies, however, as with most companies in general, do not have the kind of brand recognition that eBay does. You probably don’t have to remind scientists that Sigma sells chemicals or that Illumina sells sequencers, but these are the exceptions rather than the rules. So what’s the takeaway for smaller companies? First, while search engine marketing is a fantastic tool and can work wonders for sales or lead generation, we shouldn’t simply expect it to do so. Secondly, testing and analytics are extremely important – not just for search marketing but any advertising campaign and most marketing endeavors. While it may be more difficult to draw accurate conclusions from smaller sample sizes, most of the experiments that eBay ran to test their hypotheses could be done by any company.