The provider of cloud-based e-commerce and business operations software says it’s getting strong demand for its software from B2B companies as well as retailers, ...
Analytics help retailers sift search traffic to find the sparkling prospects for search optimization.
Retailer Vintage Tub & Bath sells nearly 150 different kitchen faucets-pot-fillers, pumps, single-lever and more. It’s a huge selection, but it doesn’t include every brand. It doesn’t sell Grohe kitchen faucets, for instance.
Vintage Tub & Bath could optimize a page-or more than one-for the broad search term “kitchen faucet” and bring in lots of traffic from natural search. But if visitors are looking for Grohe kitchen faucets, they’ll likely visit a page or two and then exit. Take that scenario, multiply it by “antique kitchen faucets,” “builder’s surplus kitchen faucets” and any other kind of kitchen faucets Vintage Tub doesn’t happen to sell, and it’s easy to see why the retailer doesn’t waste its resources optimizing pages for the term “kitchen faucets.”
Vintage Tub and most other retailers now have enough experience with traffic from search engines to know that volume doesn’t necessarily add up to revenue or profits. Early on, they learned that analytics packages could identify the paid search keywords that lead to the most sales.
Digging more deeply
Now they’re digging more deeply into analytics data to identify not-so-obvious metrics that can become action points for optimizing their pages for natural search-with more such metrics being developed all the time.
Uncovering additional new connections between search and shopper behavior can yield insights that allow retailers to drive stronger results from search engine optimization. The additional data can help them better structure their pages so that search engine crawlers will deem the content relevant and serve a retailer’s listing higher in search results.
And the application of analytics to search results helps retailers avoid wasting precious resources. For natural search optimization doesn’t come for free. It requires retailers to spend staff resources to change title tags, page content and implement all the other tactics of on-page optimization.
Though he estimates fewer than 10% of retailers do so, Chris Knoch, principal consultant in the best practices group at analytics vendor Adobe Omniture, says retailers can determine a hard rate of return on keywords optimized for natural search much as they can for paid search. They do that by adding up the cost of internal resources or outside agencies hired to conduct optimization across the entire SEO program and calculating return against that. “That gives them an ability to extrapolate a cost per click, even for SEO,” Knoch says.
Paid illuminates natural
Vintage Tub & Bath, for example, recently delved deeper into the data from its Adobe Omniture software to study a connection it hadn’t studied before. It’s measuring how often a keyword it bids on for paid search ads delivers visitors who then take an action that moves them closer to a completed purchase, such as viewing a product or adding it to a cart.
While Keyword A may generate 100 clicks that result in 221 product views, Keyword B may generate 70 clicks resulting in 547 product views, says Internet marketing manager Mike Deckman. “That’s the keyword we would target for optimization. Even though Keyword A sends more traffic, Keyword B sends more qualified traffic which would result in more overall revenue,” he says.
Omniture calls each step toward a purchase a “micro-conversion” and Vintage Tub for the past six months has been using that metric to evaluate the keywords it bids on for paid search ads. Paid search is a quicker way to test whether a term drives qualified traffic than natural search optimization, as it can take a while for improvements to site pages to move a retailer up in organic search results. With paid search, by contrast, the retailer can measure each shopper’s actions and tie them back to the keyword.
Vintage Tub is now using what it’s learned about which keywords produced high levels of micro-conversions to prioritize the search terms to optimize for, and which landing pages to focus on. Dick won’t provide details on this approach but calls results “favorable.”
The ever-popular ‘We’
Besides identifying keywords that set many consumers on the path to purchase, analytics can also help retailers go after revenue by identifying keywords that the retailer might not think of. In part, that’s because retailers tend to be familiar with their own products. Thus, the average consumer electronics marketer probably knows as well as her kids’ names how to spell Wii, the popular Nintendo game console. But a consumer looking for a gift for her grandson might search for the term “We” instead of “Wii.”
Online marketers should mine their analytics data to identify terms important to them that are often misspelled, so they can produce content that resonates with online shoppers, says John Squire, chief strategy officer of web analytics vendor Coremetrics. “My favorite example is a debt consolidation company whose number one converting keyword was ‘debt’ spelled ‘det,’” Squire says.
In cases like this, a company can still claim the traffic from a common misspelling of a search term by including it in a page’s title tag or URL when optimizing for natural search the misspelling along with the current spelling. A web crawler could see it in the title tag or URL, and a search engine could serve up a search listing that displays that incorrect spelling, taken from the title tag or URL. That way, a search for “det” consolidation could turn up a web page well-optimized for “debt” consolidation, with the word spelled correctly in the text that’s visible on the page.
Analytics firm Coremetrics notes that Google and other search engines are designed to recognize frequently occurring misspellings of keywords, and over time they start to suggest the correct spelling at the top of the results in the form of a question to the searcher-“Did you mean XXX?” As that occurs, it reduces the need for retailers to take such corrective measures in optimizing pages for natural search.