Sponsored Content WEBINAR: ELECTION 2008: E-mail Tips from a Political Insider   WEBINAR: Becoming the Illusionist: The Magic of Using "Brand" to Drive Response without Results Disappearing in the Process   WEBINAR: Score and Win Through Cooperation   TOOLS OF THE TRADE: Holiday's are prime-time for email marketing - maximize your ROI by getting to the inbox   ON DEMAND WEBINAR--MASTER CLASS: High Performance Online and Off-line Strategies for Large Scale Customer Acquisition   FREE WHITE PAPER - Direct Mail and the Internet Find any supplier you need - agencies, CRM, fulfillment, lists, e-commerce, paper, printers, telemarketing, and more. Featured Categories Lists and Data Telemarketing Database Marketing E-commerce Web Marketing Agency & Creative Services Print, Production & Paper Lists and Data Processing :: view all categories Get free access to more than 50,000 list data cards - one of the most comprehensive databases in the industry. >> Search Now DIRECT Buzz Blog Webinars Buyer's Guide Calculators JobZone ListFinder Promo Calendar Research Store E-Newsletters Print Magazine This Month in Direct Magazine Bare Bones Postal reform works. The rate hike that takes effect May 12 is the lowest in memory: an average of 2.88%. And some mail classes are getting even less than that... See Full April Issue

You know you should be optimizing for natural search. Eye-tracking studies of user behavior say that most people spend most of their time looking at natural results rather than paid search ads. And rising keyword prices have made it more cost-effective than ever to spend your time and resources lifting your site’s position in the organic rankings on Google, Yahoo! or Windows Live Search.

But knowing that something can be a help is a far cry from being able to accurately estimate the benefit your site can get from thorough search engine optimization (SEO). Visualizing that boost is particularly difficult in the case of large Web retailer sites.

Madison WI-based interactive marketing firm NetConcepts set out last year to find a way of projecting what SEO can do for the average online retailer. Specifically, the agency hoped to evolve a formula for estimating the upside of optimizing retail sites for the long tail of unbranded keywords that most of these sites have available to them.

“Many retailers are driving traffic through paid search channels, and they have a gut feeling that natural search should also be working for them,” says Brian Klais, NetConcepts vice president of search and an author of the report produced from the study, “Chasing the Long Tail of Natural Search”.

“What tends to happen is that merchants do ten simple things to their sites, optimizing their URLs and adding metatags to their pages,” he says. “But they never really have a good quantitative projection of what stronger natural search can do for them.”

There’s a good reason for that. Producing a highly specific analysis of the potential benefit from SEO on a given retail site would mean adding up all the words on every page and researching how many users search on each of those words. Web operators would also have to calculate the likelihood of each of those pages ranking high enough on each search engine to get significant traffic for those terms.

Klais and NetConcepts set out instead to develop a general rule of thumb for producing such estimates and some useful benchmarks for measuring the natural-search performance capabilities of large e-commerce sites.

In particular, Klais says, they were interested in the impact of optimization on unbranded keywords. Most e-commerce merchants are savvy enough to optimize for their branded terms, either their trademarked name, their .com URL or name plus a product (“petsmart dog collar”). But this is largely a defensive tactic, something every competitor does to guard online market share. Buyers who are actively researching a product for purchase are much more likely to enter the product name or model without linking it to a retailer’s brand.

For that reason, the company examined 1.2 million unbranded natural search visits to the sites of 25 of the company’s Web retailer clients, with a total of around 5 million Web pages among them. The sample consisted of both highly popular and more specialized sites from a range of consumer categories. Over half of the sites were ranked in the Multichannel Merchant 100 list of top online retailers.

Sure enough, a scrutiny of the natural search traffic those clients drove found that 80% of it came from branded keywords. It also revealed that only a small percentage of the unique pages on those retailer Web sites actually drew natural search traffic—on average fewer than 5% of a merchant’s total pages, and mostly through the deployment of those branded keywords.

“Large retail Web sites can have tens of thousands of unique dynamic pages, with SKUs, products, subcategories, categories, departments, and so on,” Klais says. “All of those should be getting crawled and driving traffic from Google, MSN and Yahoo! Why in the world have you got those thousands of pages if most of your traffic comes from your company name? And how much opportunity are you missing out on?”

“As it turns out, the missed opportunity [for unbranded natural search optimization] is huge. For most large Web retailers, the bulk of their site pages are freeloaders, producing no search traffic.”

The white paper goes into some detail about the calculations behind its formula for estimating how many non-branded natural searches a large and complex retail site should produce, a concept NetConcepts calls “Page Yield Theory”. Those steps can’t be easily summarized, and anyone interested in the equations underlying the rule is best advised to download the study at http://www.netconcepts.com/long-tail-of-natural-search/.

Basically, though, NetConcepts found that:

* Only 14% of the group’s indexed pages yielded natural search traffic, leaving 86% of pages with unrealized potential.

* On average, each site studied contained 2.4 unbranded keywords per page, and each of those keywords produced 1.9 search visits, for a ratio of approximately 4.6 keyword visits per unique page.

* The group saw 189,000 brand searches conducted in January 2006.

* Checking reported traffic on a sampling of keywords to clicks received by the merchants in the study, NetConcepts calculated a 4.7% clickthrough rate for the average keyword.

Using those standards, Klais and company determined that the potential search traffic for a large Web site will be about 97 times the number of unique pages on the site. Anything less than that counts as SEO underperformance, to some degree.

At 97 searches per unique page, the client group under study had the potential to produce 7.1 million keyword searches a month. Since the sites were on average adequately optimized for brand searches, that unrealized potential lay buried mostly in unbranded keywords.

In other words, the sites studied could have been receiving unbranded natural search traffic at a rate 38 times the 189,000 branded searches they got in the month they were studied.

Marketers who aren’t equipped to calculate their keywords per page or their visits per keyword can use a rough guide from the NetConcepts model to figure out how many of their pages are yielding up natural search traffic. The merchants in the study saw 80% of their organic traffic from branded search and were getting traffic from 14% of their pages. By those measures, a retailer whose organic traffic is 95% branded is only getting that traffic from 3% of site pages. To achieve a 50-50 branded/ unbranded split, that merchant would have to get search traffic from 58% of his or her Web pages.

Klais says the NetConcepts findings can lead to a new way of thinking about brand-building on the Web.

“If I’m searching Google for ‘furniture slip covers’, I may not know that L.L. Bean or Wards.com sells them,” he says. “What’s it worth to those companies to be associated with those terms in natural search? There’s a real opportunity here for Web retailers that have strong brands to leverage both the large size of their Web site and the strength of their brand to win those unbranded searchers and convert them into clicks.”

But even with better insight into what they stand to gain from optimizing for non-branded searches, large-site Web retailers are still faced with the problem of how to get started. Taking a single Web page that now gets two search visits and optimizing it to get four isn’t an efficient use of resources, especially for a 50,000-page Web site. Rather, merchants need to look for scalable tactics that let them cover a lot of ground with little effort. Klais has some suggestions for improving natural search optimization without undue pain:

* Change page URLs from complex and dynamic to simple and static ones. This can have a big impact on search traffic, not to mention enhancing a site’s crawlability by search bots.

* Take a “thin slice” approach to optimizing copy. Look for small changes that can be made on lots of pages in a short time—for example, improving title tags. Then watch to see where these changes have the most effect on search optimization and expand your efforts in those places, making further revisions and monitoring those results too.

* Get your customers to help improve your natural search rankings. This can be done by incorporating consumer-based content, letting visitors tag a page in their own vocabulary, offering RSS feeds and blogs as link bait, and providing lists of most-popular keywords others have used to find a page.