Want to learn to improve your customers' experience, pinpoint buyers' exact interests and reach prospects at every phase of the buying cycle? Look to your search logs to mine this crucial market research data.
A special report from the Search Engine Strategies conference, August 2-5, 2004, San Jose, CA.
During the session "Using Search for Market Research" at Search Engine Strategies San Jose, industry experts Ron Belanger from Carat Interactive and Kohn Klimstra from iTraffic discussed how to transform data details into behavioral marketing gold. Rather than discussing the merits of a page-one versus a page-two listing, the pair stressed that data mining helps Web marketers understand their customers' motivations and needs.
"[Marketers” need to think beyond rankings and traffic and think about why someone buys," says Belanger. "The more data you collect, the more you can mine for post-sale analysis."
What are your customers searching for?
Belanger launched the discussion by reminding the audience that the first place marketers can review raw data is through search queries. This "pre-click" behavior helps site owners construct search campaigns that target their customers' interests and extend a company's online branding.
"Determine what percentage of search queries are brand-related and what keywords are related to your brand," says Belanger. "Then ask, 'Currently people associate our brand with what products and services?'"
Belanger recommended companies evaluate their keyword qualifiers (such as the word "cheap"). Although "cheap" may be effective for Best Buy it would not be effective for a luxury brand like Mercedes.
"It's important to know what adjectives are used for your products and services," says Belanger.
Klimstra added that reviewing customers' search queries helps pinpoint where they are in the buying cycle. Then, site owners can serve content targeted towards those queries.
Klimstra presented a financial services company case study where the campaign goal was to generate home equity loan applications. Digging into the data demonstrated that prospects who used what Klimstra called "benefit keyphrases" like "debt refinance" were searching for educational information about refinancing their home. People searching for "product keyphrases" such as "home equity loans" were in a more advanced purchasing phase.
By targeting both "benefit" and "product" keyphrases in their search marketing campaign, they were able to reach people at all phases of the "conversion funnel" from initial research to eventual purchase.
"You need to ask what consumers are interested in," says Klimstra. "How are they searching?"
Combating search marketing "conversion-itis"What if a landing page sees sluggish conversions, and customers click away rather than buy? Or a keyterm isn't driving traffic that makes the hit counters click? Does it mean that the keyterm or page should be tossed on the search marketing landfill? Both Belanger and Klimstra argue that a campaign's greatest failures can provide the most intriguing - and possibly profitable - data.
Marketers rely on conversion metrics to determine a campaign's effectiveness. If a campaign is not performing adequately, marketers typically discontinue terms without reviewing the data first.
"We tend to cut our failures rather than learn from them," stated Klimstra.
According to Belanger, marketers should move away from a conversion-only focus. By stressing conversion at the expense of information, marketers may not learn about their audience.
"A lost sale does not mean that you are targeting the wrong audience," Belanger says.
For example, low landing page conversions may not mean that a site owner is driving the wrong traffic. The page may require change in messaging or a different landing page layout - and making a few simple tweaks will jumpstart sales. Or, customers may research a product or service online - but purchase offline.
"Search marketing is driving offline sales, but we aren't getting credit for it," Belanger says.
If a keyphrase or campaign is not performing up to par, Belanger recommends that marketers mine those lost sales for additional information. Armed with more complete data, site owners can determine how a campaign is actually performing and learn how to more effectively reach their customers.
Leveraging landing pages with customer-centered communication
Klimstra stressed that marketers should continually review their data and determine if their messaging is hitting the mark, or not connecting with customers.
"You must look at click rates and determine how well your site interaction meets [your customer's” needs," says Klimstra.
Klimstra provided a case study from Barrie Pace, a women's designer clothing store. During a special traffic drive targeted to mothers of the bride, the campaign saw over 25,000 visits, but experienced a low conversion rate. Rather than discontinuing the campaign, iTraffic reviewed the data and visited competing sites. In the process, Klimstra realized that the messaging wasn't focused towards the mother of the bride.
Their revised strategy involved creating a special landing page for mother-of-the- bride fashions. The page was customized exclusively towards the mother, including copy that provided a "Mother of the Bride Tip." After revising the landing page, conversion rates doubled. Klimstra credits the customized landing page with the higher conversion rates.
"Landing pages targeted towards keywords help customers determine if they are in the right place," says Klimstra.
Data mining tips and toolsDuring the session Belanger and Klimstra provided some additional data mining tips, including:
- Tracking search marketing's effect on offline conversions is difficult for many marketers. To combat this problem, Belanger suggested installing a special toll-free phone number for Website orders. If a call comes through on the exclusive line you'll know it's a lead from the web site.
- Review your site search logs. According to Klimstra, a company's search logs reveal customers' main product and service preferences. This information can then be used for promotional and merchandising decisions.
"Companies can determine how to merchandise things on the home page depending on the data from their site search," says Belanger.
Additionally, search logs can also help determine any architectural site challenges. That is, if people have to search for a product or service, it may be because they can't find it through your site's navigation.
- If you need long-term or seasonal keyphrase data, some "crude but better-than-nothing" tools are the Google Zeitgiest, which reveals search patterns and trends back to 2001, and Shopping.com's Consumer Demand Index, which shows weekly shopping search queries and patterns.
- Are customers buying but then returning products? Belanger suggests that search logs can help predict post-conversion behavior. By reviewing keyword data, marketers may find that certain keyword queries are associated with a higher product return rate.
Detailed data mining helps site owners interact more effectively with their customers. By reviewing raw data, such as site search logs and search queries, marketers can maximize their marketing campaigns and provide targeted product offerings. The more data that is measured and quantified, the more that companies can provide their customers the exact product, service or information they need, when they need it.
"You can be the next superstar of your company by just mining your data and giving customers what they are demanding," says Belanger.