Merchants know how to rank high in paid search results. But the power of popular informational sites means they have to adjust their goals when it comes to natural search rankings.
By Jonathon Love
Given the billions of dollars web retailers spend on paid search each year and their growing investment in site optimization, retailers by now should have a pretty firm grasp of search engine marketing. But a new nine-week study by Internet Retailer for the just-released Top 500 Guide of more than 300 keywords in 20 merchandising categories on Google reveals that retail web sites have a split approach to search engine marketing. While they understand how to manage and bid on pay-per-click keywords, they fall behind informational and other non-retail sites when it comes to natural search results.
In fact, in natural search results, retailers occupied fewer than half of the three top spots in the 20 categories—29 out of a possible 60 positions. Amazon.com Inc., the largest online retailer, led the industry, appearing 11 times in the 60 spots. In six of those, Amazon was No. 1. The next most frequently appearing retailer was Overstock.com Inc., a prime competitor to Amazon and No. 30 among Internet retailers, with three. No other retailer showed up more than once in the 60 possible spots.
Appearing in natural search results is key to a web site’s success. Several studies have shown that consumers’ eyes jump over paid results and go direct to natural. For instance, on Google, one study showed consumers preferring natural results 3 to 1.
Though to retailers, search ranking is a complex, unpredictable process, Google and other search engines work hard to rank what they deem to be the correct sites. If they view a query to be 60% informational, for example, six of the top ten results will be informational, some experts say. If they believe a query to be largely transactional, most of the results will be retailers or comparison sites.
Informational sites stand to outrank retailers in natural search in almost every case, especially for relatively general keywords like the ones used for this research, because they have the capacity for seemingly unlimited content and they can overwhelm retailers in terms of link popularity. While it’s correct to assume that user-driven sites like Wikipedia might not be the most credible in terms of information accuracy, the credibility they have generated comes from the extreme popularity they have achieved. “I don’t think retailers can ever surpass Wikipedia because of the two key pillars in search engine optimization—relevant content and sites linking to you, both of which Wikipedia dominates,” says Suzy Sandberg, president of PM Digital, a search engine marketing company.
But while retailers may never be able to top Wikipedia and other informational sites in all categories of keywords, they can take steps to boost their sites higher in search engine listings.
“Whenever a retailer sees a Wikipedia listing appear in search results, that retailer should immediately ask: ‘Are there pages on my web site that meet the searchers’ informational needs?’” says Shari Thurow, founder and search optimization director at consultants Omni Marketing Interactive. “Many retail web sites contain informational pages, but many retailers might not realize it.”
There are several types of informational pages that can assist in search engine recognition for informational queries: buyer’s guides, blogs and how-to pages are all good examples, Thurow says. An appliance retailer, for instance, might add a how-to page describing the selecting of a proper dishwasher, or a home furnishings retail site could insert a page describing the assembly of a shelving unit.
Creating long-lasting content
Creating such informational pages addresses one of the biggest problems of landing high in natural-language search results: long-lasting, static content. With their ever-changing inventories, retailers rarely have content on their sites long enough to make search engines believe that they should rank high in search results for a long time. “Retailers sell inventory, and inventory fluctuates,” says Robert J. Murray, president of search engine marketing company iProspect.com Inc.
Apart from such content, a further challenge retailers face is that of links. Search engines give higher credibility to sites that have a lot of links, something that is tough for most retailers to accomplish. Retailers can address that issue, in part, by using so-called “bread crumbs,” Murray says. Bread crumbs, or the trail that usually appears at the top of a page that indicates where a shopper is in a site’s hierarchy, create a navigational outline that shows a shopper who has entered a product page from a search engine how to take steps backwards in the site’s hierarchy to a category page. This type of navigational tree creates interlinking and goes a long way, Murray says, in terms of search engine recognition.
Another important tip for retailers is to optimize category pages to accommodate the plural form of their targeted keywords because searchers are looking for a list when a query is plural, and category/channel pages often contain product lists, Thurow says.
Apart from particular steps that retailers can take, they should also look closely at their own site optimization efforts. A three-year study by search engine optimization company Oneupweb of Top 100 retailers’ optimization efforts confirms that they’re not very good at it. The most recent study reports that only 40% operated well or moderately optimized sites. The study looked at site architecture, meta tags, keywords, content and other factors affecting the site’s ability to be indexed by search engine spiders and to be well-positioned in natural search results.
The process may be more difficult, but spending to optimize a web site pays off in the end, Murray says. His clients that have had the most success at optimization and tracking results are those who understand the concept of creating flat, static content.
Who’s where?
So who’s occupying those spaces that retailers are not? Wikipedia.org, the online collaborative encyclopedia, held 17 of the spots, including 11 No. 1 spots. Other information providers, comparison shopping sites or shopping portals occupied 12 spots, with About Inc. holding three of those. The two remaining spots not occupied by retailers were held by Microsoft Corp. and the U.S. Department of Energy, both in the Home Improvements category.
The Internet Retailer study analyzed the results of 14 to 18 keywords for each merchandising category. Informational and other sites dominated some categories that seem natural for retailers to appear high in. In appliances, for instance, Wikipedia, epinions.com and ConsumerSearch.com occupy the top three listings. Sears Holdings Corp.’s Sears.com ranked sixth. Best Buy Co. Inc.’s BestBuy.com, which offers an online kitchen and laundry design center that helps shoppers determine which appliances they want, ranked No. 17, even behind NASA.
In consumer electronics, Circuit City Stores Inc.’s CircuitCity.com, Overstock.com and Crutchfield Corp. ranked fifth, seventh and eighth, respectively, behind Wikipedia, CNET Networks Inc., Amazon and HowStuffWorks.com. But BestBuy.com ranked No. 24 right behind TigerDirect.com, an e-commerce site operated by Systemax Inc., at 23.
In pay-per-click search marketing, the story is just the opposite. Retailers occupied all but 11 of the top 60 spots. In paid search categories, Sears and Target Corp. each held four of the top spots, Amazon, Circuit City, J.C. Penney Co. Inc., Lowe’s Cos. Inc. and Staples Inc. each held two. No other retailer was represented more than once.
A measure of how well retailers have nailed down paid search marketing is reflected in how Internet Retailer scored retailers when they appeared in paid rankings. To rank web sites within their merchandising segment and by keyword, Internet Retailer assigned point values every time a retailer appeared in the first five paid and natural results in each category. For instance, that total gave Sears.com in the appliance category 384 points and No. 2 Lowes 179 points. The corresponding scores for natural search results in appliances were 90 for Sears (223 for Wikipedia) and 26 for Lowe’s, which ranked No. 22.
More measurability
Retailers are more comfortable spending their money advertising in paid search listings because results are easier to track, Sandberg says. It is much simpler to advertise on a pay-per-click basis and see the return for every dollar spent than to sink a substantial amount into re-tooling a web site to make it more search-engine friendly, with longer-term results.
The fact some retailers are submitting winning bids for multiple terms within a category—in some cases as many as 50% of words in that category—tells search engine marketing analysts that online merchants are using a broader mix of paid keywords and phrases to drive traffic and sales, build brand awareness, and prevent competitors from bidding on certain words.
The research turned up some interesting anecdotes of retailers bidding on both very diverse and similar sounding paid keywords. For example, Victoria’s Secret, a multi-channel retailer of underwear and intimate apparel, was the top bidder over nine weeks for the keyword “outerwear” in the apparel category.
Another good example was “couch purse,” a paid listing by Coach Inc., the handbag and leather goods company. The company explains that “couch purse” is an intentional buy that assumes that shoppers sometimes misspell what they’re looking for. In this case, shoppers inadvertently type “couch” when they mean “coach.” Coach never displays links showing “couch” alone—it’s always with “purse” or “handbag” or other Coach products. And the link takes shoppers to the Coach page that displays the appropriate product. The gambit pays off, Coach says: The conversion rate on those terms is about the same as the conversion rates on legitimate Coach terms.
The research shows retailers are using bid management tools and keyword vocabularies to bid on terms in their core merchandising category. But appearing high up in the paid listings on Google isn’t the only way retailers can generate a better return on pay-per-click marketing. They can also do a better job of testing keywords and scrutinizing the economics of their keyword spending, says Alan Rimm-Kaufman, president and chief technology officer of direct marketing services at consulting firm The Rimm-Kaufman Group.
The importance of testing
Retailers should build up and test a keyword inventory that relates to every SKU and product in their online inventory, he says. “Retailers test their campaigns and measure the results, but they should also compile five to 10 keywords for each of their SKUs and test those,” Rimm-Kaufman says. “The additional testing will produce better keyword combinations and more focused campaigns.”
Beyond measuring the average conversion rate and traffic generated by a pay-per-click campaign, retailers should measure other metrics such as results by concept, phrase, ad copy, landing page, specific search engines and even time of day.
“Retailers can see who their competition is by viewing the rankings on a Google page,” Rimm-Kaufman says. “But they shouldn’t weigh the total outcome of a paid campaign simply on where they rank on a page. They should focus on whether the campaign generated the best return on investment by looking at all of the relevant points.”
With so many relevant points in search optimization and paid search campaigns, retailers have a lot of analyzing to do.
jon@verticalwebmedia.com
Click Here for the Search Engine Marketing Products & Services Guide
