Where Is Search Going: Yahoo’s Shashi Seth
I’m continuing my look at what search might look like in the future with this first part of a fascinating interview with Shashi Seth, the Senior VP of Search Products at Yahoo. With the recent Bing arrangement, Yahoo now finds themselves in an interesting position in the search game. Now that Yahoo search results are powered by Bing’s algorithm, the company can move its focus from maintaining a web index to how the information is presented to the user. It’s analogous to a car manufacturer farming out the drive train assembly to a partner so they can focus on body styles and interiors. So, my first question for Shashi is now, in the new search ecosystem, what does the Senior VP of Search do at Yahoo?
Seth: Our goal with Yahoo Search was to rethink search, essentially. The reason I came to Yahoo was because I felt that search had been fairly stagnant in the past 5-6 years and hadn’t evolved a significant amount in that timeframe. As you have alluded to in your columns, I felt that there’s a lot of innovation that had been sitting on the sidelines waiting for some major event to happen. I think with the launch of Bing, a little bit of that kind of innovation came about and various search engines responded to it, but I just felt that largely search needed to go through a much larger shift in the coming years. Users were demanding it and search engines had not been responding to it.
I think our mission is to rethink and reimagine search entirely. Given this relationship with Microsoft, we are shifting our focus from the very traditional indexing, crawling, and ranking to what we define as the next generation or the new face of search. That includes new ways to search for information, new ways for information to be brought to the user, new sources of data, and so on. Search as we know it is going to change significantly.
I asked Shashi what this meant: Is Yahoo’s focus squarely on the user experience now that Bing is powering the algorithm?
Seth: That’s correct. The one thing I would say is that we use the word “experience” or “user experience” loosely, and I’d say that it’s not just the user experience. The kind of data that comes or is presented to people is also very different. I think in the new world, for example, people are much more inclined to get the exact answer when one exists rather than a host of different information that we present to the user and hoping that they will dive deeper into URLs, for example, to find that information.
This brought up a limitation I’ve found with the current state of search. As users we have master intents—for example buying a new car or moving to a new neighborhood. But within that master intent are several different tasks we have to accomplish. A move might entail looking for a home, looking at schools, finding a mover and researching a new community. Each of those subtasks could further entail dozens of searches. But then, it’s up to us to piece it all together to meet our original intent. A truly smart engine should be able to decipher our master intent and help us through this process. This point came up in my interview with John Battelle. So, I asked Seth, are engines starting to get that smart?
Seth: In some ways it has started happening in the back end. Let’s take a look at what happens to a typical query. So when a query comes in, the first thing that we do is obviously go through our index and look up all the results that are relevant, and in doing so we also use various forms to rewrite the query—for example, use of synonyms, use of different keywords that are juxtaposed in order to either expand or contract the query, and so on. All those things are being used by all the search engines. But at the same time when that query is fired off to our back end, we also do what we call “look aside” indexes. So, for example, if we don’t know for sure that this could have any local relevance, we may fire it off to our “look aside” index and be able to see if there are meaningful answers, and if there are, then we might choose to paint the search result page in a different way then if there aren’t any.
There’s a lot of stuff that happens in the background that today gets to finding what the master intent was, and that guessing today is done by what we get back from various indexes, including the “look aside” indexes. So “Is this query of local intent?” It could be if we get back a lot of very high-quality, relevant results from the “look aside” local index. At that point, even though we didn’t know right at the beginning that this was a query that had a lot of local intent in it, we might surface it that way based on what we get back. So in some ways I think what search engines have been doing is learning over a period of time what the master intent is and doing it a little backwards in order to get to the answer.
And then, Seth started to go down one of my favorite paths: the impact of mobile and searching through apps.
Seth: But interestingly there’s another phenomena that is happening, and that is with the advent of mobile search and the use of apps as a proxy for search. I think people have narrowed down that master definition or master intent up front by choosing the app that they choose to engage with. So, for example, if you are looking for local results, you may use a very appropriate local app in order to do those local searches, and therefore there is no guessing to be done anymore. By nature of what you just did, we already know that you are looking for local results and that is the only index that I have. When somebody uses a mobile app and we know what the GPS coordinates are and the fact that they’re using a local app.
But this brings up a thorny problem. If our searching fragments over multiple platforms and apps, so does the opportunity to present relevant advertising. We no longer have a universal search destination. And that poses a much more difficult monetization challenge. You have more signals of intent, so you can serve move relevant ads, but you have a significantly more complex ecosystem you have to consider in serving the ads. No longer does Google (or any engine) have a virtual monopoly on the eyeballs searching for information. What does this mean for search advertising?
Seth: I’ll put a couple of stakes in the ground and then you can draw from that. Number one, I believe that in five years or so, mobile search will have overtaken desktop search. And the areas where that is going to grow the fastest is going to be in Asia and Europe. I think US will follow a similar curve, but like most traditional mobile expansions, North America will be a couple of years behind those two markets. The current interesting thing about that is that in Asia, brand advertising and search advertising are pretty much nonexistent. If you compare the revenue numbers from North America or Europe versus Asia, it shows there hasn’t been a large acceptance of those kind of ads in those marketplaces.
But the interesting thing is that given the maturity of the mobile acceptance in that geography, that mobile advertising in many shapes and forms already exists. It may not be as scalable as we’ve made search advertising, but definitely that mindset—and the dollars—exist. In the next 3 to 5 years I wouldn’t be surprised if mobile advertising, including search, leapfrogs those two traditional forms of advertising and completely changes the game.
Now we’re talking a significant disruption in the placid waters of search marketing. And any time there’s disruption, there’s an opportunity for new players to grab market share from established players. In the search game, that means grabbing share from Google. As search becomes increasingly mobile, we have to consider not only the usual suspects—Google, Yahoo and Microsoft—but also a dark horse like Apple.
Seth: Absolutely. I completely agree. And I would say that with the mobile advertising also comes another opportunity, and that is to break down the artificial boundary between search advertising and display advertising. And that’s another stake in the ground, which is that mobile search is not going to look and act much like desktop search. In fact, I believe that mobile search is going to be in the shape and form of mobile apps and that users are going to engage with it and therefore the kind of advertising that lives there is unlikely to be traditional search advertising.
But it’s not just advertising that is at risk for disruption. The fact is, a search experience on a mobile device is a new environment, and it will create new expectations. When expectations shift, habits can be broken. And, as I’ve said before, the use of Google for desktop search is a habit. Searching through apps brings a new focus to the experience, which allows for deeper functionality and cleaner presentations of results. The restriction of real estate on the mobile device forces search to get right to the point, and that could alter the experience on our desktops as well.
Seth: I completely agree with that hypothesis. In fact, we’re starting to rethink some of our experiences on the desktop search side and start bringing in a bunch of things that we’ve learned on the mobile app side.
My personal belief is that the changes we’re seeing on Google recently is their attempt to start creating a differentiated experience on the desktop and on mobile devices. Search is getting ready for the user experience divide that will be ushered in by the adoption of new devices. This will be one of the topics I cover with Shashi in my next Just Behave post.
Opinions expressed in this article are those of the guest author and not necessarily Search Engine Land. Staff authors are listed here.