NAA Day 2: Local Search
The highlight of Day 2 of the NAA show was a local search panel. The length of sessions (90 minutes) results in a lower number of sessions than I’m used to attending, but the quality has in most cases made up for the quantity.
Local search has many applications to newspapers in the many forms it takes, including directory listings, classifieds, video, and mobile. Integrating all these elements to create more robust and attractive local destinations was a core theme of this panel.
Google’s education of the marketplace on what search should be and the bar it raised in doing so has made this job more difficult for traditionally tech-averse newspaper publishers (the industry’s overall place in the tech learning curve could be gleaned from the explanations sometimes required from the stage when terms like RSS were used).
But as mentioned yesterday, the industry is coming around and has made some sizable steps in some cases. There was a time in the not too distant past when some newspaper execs were liable to think that VoIP is a low-carb vodka and blogging is a primitive form of punishment (sometimes it is). But there is still a long way to go and, to their credit, many in the industry realize this at least that’s the message that could be heard clearly all week here in Las Vegas.
So users’ expectations of the search experience have been molded by Google, and newspapers have a significant challenge in meeting that expectation. This isn’t only due to a lack of ability or affinity for the technology required to make it happen but is also, as I’m beginning to learn, a function of the discreet and segmented nature of newspaper content.
In order to create unified search results that combine news, classifieds and Yellow Pages content (which is something to shoot for, as we’ve said in the past), these traditionally disparate buckets of info must be brought together. There are good reasons that they erstwhile have not, besides newspapers’ lack of online innovation, which is often blamed for better or worse.
The challenge is that these different types of content have different data formats and there are correspondingly different ways to index them. Similarly, users use different terms and different search behavior to find each of them. Search queries in news can be myriad in quality and quantity when compared with that of classifieds. Classifieds in turn involve broader and more search terms than Yellow Pages content, which, by comparison, has content neatly packaged into a much smaller bucket of standard search headings.
Oodle’s Craig Donato offered this “reality check” during a classifieds/directory convergence panel at the last Drilling Down on Local conference, and Terry Millard of Planet Discover expressed similar thoughts at the NAA show.
“In automating unified search, you will get roughly 60 [percent] to 80 percent accuracy in contextually relevant results,” he said. To improve this, many on the panel agreed that an investment in manpower is necessary to manage the accuracy of the data and contextual relevance of search results. The upfront requirement to make this happen is difficult to execute and has proved scary for many publishers, according to Millard.
Another option is to integrate “social search” platforms that utilize users to tag content. This can involve a difficult adoption curve, which is why some private-label social search platforms available to newspapers run a “passive tracking” of user behavior to identify their interests (specific to a certain site, user or grouping of users), without requiring them to actively do so. These include companies we’ve written about here such as Collarity and Eurekster.
Many other challenges come in to this discussion such as branding (using the legacy brand or a new domain a question of “brand baggage” vs. brand equity). The investment and patience required to face all these challenges and create a compelling local search destination was continually echoed by others on the panel (rest of panel members listed here).
After all, Google wasn’t built in a day; it took four years to monetize its content and see ad revenues. Unfortunately many newspapers don’t have that luxury because of finite resources and falling revenues. But it shouldn’t take that long, and opportunities do exist despite all these challenges.
“It won’t take four years, but it also won’t happen overnight,” asserted Millard.
Tomorrow, Yahoo!’s Hilary Schneider will have a much anticipated keynote that I’ll cover here along with more of a conference recap, and key takeaways from other sessions. More on today’s events can also be found over at Search Engine Watch.
