This week at SMX West you can see members of the Live Search team speaking on a range of search topics.
I’ll be there as part of a panel on search user behavior called “Just Behave – A Look at Searcher Behavior”, at 3pm on Wednesday the 11th. It should be a fun session, with some great people from Yahoo, Comscore, and Enquiro.
My primary job at Live Search is Group Program Manager for Search Relevance – making sure the results we return for your query are the right ones. It’s a fun job with a lot of unique challenges, great people, and uniquely cool technology. It’s fun to be in the race as an underdog. At the same time, I’m the kind of guy who is always asking “why”. The “why” of a search engine is to help people find the information they’re looking for as quickly, easily, and confidently as possible. If you want to do a good job at that, you need to understand the people you’re trying to serve and the context in which they’re using your tool.
We’ve learned a lot over the last few years about how customers behave with search engines. But there is one key insight that really sticks out in my mind that I’ll be talking about at SMX: searches don’t occur in isolation – they are often part of a longer task. That is, searchers come to a search engine with something in mind, do an initial query, click on multiple results, perform multiple follow-up searches, and then frequently come back in the following days or weeks to return to the same topic.
That’s not to say that we don’t see a lot of straightforward, one-query, one-click activity. We do see a large number of these navigational searches seeking a specific site, such as:
These types of searches are usually quick hits. The median time we see for these navigational search tasks is measured in seconds. Customers (on average) issue the query, spot the result they want, and click fast.
How searchers spend their time
While navigational queries are common and account for a fair percentage of our search volume, they represent a relatively small amount of customer time spent on a task. Most time spent on search tasks is actually spent doing the kind of research that requires visiting multiple pages – either to find exactly the right one, or because the task itself requires going to several places.
Let me show you an example. Consider this search session below (shown by Satya Nadella in his SES San Jose keynote in August 2008):
Contrary to the normal query-by-query view of search, this is actually a whole task that lasts at least half an hour and possibly closer to an hour (depending on how you interpret the time spent on www.easyspirit.com).
This task shows that customers engage in an iterative research process as they lead up to making a decision. They search, sift through options, reject ones that are obviously irrelevant, and visit pages that seem most likely to offer the information or ability to take the action that they want. But they do not necessarily stop at the first good option. Instead, they learn from the content they absorb, use that to refine their searches, and then frequently continue with their research until they reach a critical mass that gives them the confidence needed to make a decision.
What does this mean for site owners?
We think this behavior has important implications for webmasters and advertisers. This same behavior occurs whether the query is about shoes, cars, health issues, astronomy, recipes, etc. Just name a topic and we see that customers engage in long, intensive research sessions, visiting multiple pages, issuing multiple searches, and often returning to the best pages again at the end of the session or on a later date.
Our analysis shows that 46% of the time customers spend on search tasks is on tasks that last half an hour or longer.
What does that mean for webmasters? Many things, potentially, but here are two thoughts:
Well, I’ve digressed into the philosophy of how to build an e-commerce site, so I’ll draw this post to a close. Come visit the “Just Behave” session at SMX to hear more.
In a future post, I’ll talk about how customers come back to the same or similar tasks over time. And you’ll also hear more from us this year about other aspects of customer behavior, about our relevance algorithms, and about the scientific process we use internally to drive our research and development.
Until then,
Ramez Naam, Group Program Manager for Search Relevance
It seems as though people do a substantial amount of research prior to purchasing a product. Perhaps Live Search should consider entering into some form of partnership with ConsumerReports to either display an abstract or entire articles about products. This would add a huge amount of value to product searches - one category that Microsoft seems very focused on. Of course, for car research you could also link to reports from Car & Driver and Road & Track and similarly for electronics, Zdnet and Cnet.
Have you uncovered significant differences -- or implications -- for searches when they are done by business buyers exploring complex solutions? Let's say they're exploring an enterprise software category, like marketing automation or CRM, can you track where they go from site to site the way you do for consumer searches? If so, based on observed behavior, what's different about B2B searches (in product categories with big ticket items)?
Joe: Good idea on the Consumer Reports data. We will take a look at that.
Christine: We have never really sliced it by business searchers vs. consumer searchers. Interesting idea. I would say that overall, it looks like "big ticket" items may sometimes result in longer sessions leading up to the decision. But that is just a hypothesis at this point. We have a lot to learn yet. :)
Thanks
<a href=http://www.elazigaktuel.com title="Elazığ, Elazığ Haber, Elazığ İl ve İlçeleri, Elazığ Güncel, Elazığ Gündem, Elazığ Yaşam ve Sağlık, Elazığ Kültür ve Sanat, Elazığ Müzikleri, Elazığ Halk Oyunları, Elazığ Yöresel Yemekleri, Elazığ Harput, Harput, Elazığ Balakgazi">Elazığ - Elazığ Haber</a>