Confab Impressions Part I: The Search Haters
This week I attended Confab 2011 and, as I have been telling everybody, it was the best conference I have ever attended. That’s saying a lot because there was a period between 1997 and 2003 when I attended dozens of them as EIC of ComputerUser.com. Congratulations to Kristina Halvorson and the kind and cheerful staff of Brain Traffic for doing such an awesome job.
At the conference, I attended several great keynotes and other talks by such notables as the venerable Ginny Redish and the awe-inspiring Ann Handley. I have a lot of great things to say about their talks, but it will have to wait until the next two blog posts. First I have a rant about a couple of other talks I witnessed.
Two speakers at the conference exploited tired old search myths to dismiss search as a legitimate concern of content strategy. Steve Rosenbaum gave a talk about how the flood of data is overwhelming search algorithms, leaving human curators to do the work of machines in helping people find the content they need. Christine Perfetti claims that if users are forced to search for information, it is an indicator of UX failure. The arguments they use to defend these claims are flawed. In fact, as the flood of data overwhelms users, they turn increasingly to search for a short-cut to navigation. I’ll explain why after the break.
Rosenbaum’s Flawed Logic
Rosenbaum’s argument for the death of search hinged on two facts. First, since the dawn of the web, public digital information has grown exponentially and shows no signs of slowing down. It’s hard to believe any algorithm can keep up with the data deluge. According to Rosenbaum, the so-called Panda (aka Farmer) update to Google’s algorithm is a tacit acknowledgement that Google is increasingly clueless about how to rank pages. As proof of this, Rosenbaum showed two NASCAR sites that have nearly identical content: one is the first site for the given keywords, the other is on page 8 of the results.
Without showing us how the two sites are indeed identical ( keyword usage, inbound links, URL structure, metadata, etc.), the audience was forced to take his word for it. But as someone who typically sees dramatic improvements in organic search ranking from simple white-hat SEO techniques, it seems like a fluky case. He would need to do a more general study showing how multiple functionally identical pages have radically different ranking to prove his point. One case does not an argument make.
But that was not the main flaw in his logic. The main flaw was in saying search is dead largely because it is flawed. I’ll be the first to admit search engines leave a lot to be desired. Since when has it ever been much better? As information has exploded, Google’s ability to keep up has ebbed and flowed. But the general trend is toward increased search usage in spite of the data deluge. Users have learned to live with search’s limitations to do more sophisticated long-tail searches and find more relevant content.
I’m sure there is some variation between audiences, but according to Tech Target, 96 % of our target audience (B2B IT decision makers) frequently or always use search to find the information they need. According to an unpublished Google study that I have had the privilege of seeing, search usage has doubled worldwide since 2007 and the trend is accelerating. At this rate, search usage will double again within two years. It looks to me like search is alive and well.
Thing is, Rosenbaum doesn’t need to dismiss search to make his point: Curation is a growing concern among the masses. More and more people are using Twitter and other social venues to help their friends and followers find relevant information. They curate feeds of content and share them with like-minded peeps on the web. This seems indisputable. And it might cut into search’s growth a bit. But there’s just so much data, we need all the automation we can get to make sense of it all. Indeed, search is a key way savvy curators find and share relevant content to round out their feeds. What are hash tags if not devices to help users find relevant content when they search on those words in Twitter? In short, search and curation are not mutually exclusive. Rather, they complement one another.
In the course of an otherwise informative talk about usability studies, Christine Perfetti pulls out a claim that I thought was long since debunked: If a user reverts to the search box at the top of a page, it means the UX needs improvement. Study after study at IBM shows our users preferring search to navigation in many circumstances, regardless of how clear our UX options are. Since changing our masthead search function last year, we have seen twice as many users per month searching for relevant content with our internal search engine. The obvious implication is, if you give them a useful search experience, they will use it in favor of poking around in our navigation.
When you unpack this, it’s easy to see why. By way of example, Perfetti showed the UX of the Microsoft.com site, noting that the sheer volume of options overwhelms users. I would say we have the same problem on ibm.com. With literally hundreds of offerings to promote and support, there is just no way we can surface every relevant offering to users without overwhelming them with choices. In this environment, they prefer to search for content related to their informaiton tasks. There is nothing wrong with that. It is a simple fact of information overload. Besides, you can’t design experiences for every user task. Design for the top tasks and let search do the rest. And search is the ultimate top user task.
Search is the central digital behavior and it likely always will be, as information continues to grow exponentially and mobile devices proliferate. The sooner people accept this reality, the sooner they can begin designing content experiences their users will want to engage in.