Editor’s note: The following guest post is written by Shashi Seth, the senior vice president of Search products at Yahoo! Previously, Shashi worked at Google where he developed the monetization strategy for YouTube and was also the product lead for search.
Search is about to change quite radically. For more than a decade, search has been stagnant: the core product has not changed much. Users have changed radically in that time frame. Even though the kind of content users consume is different, search engines are still focused mostly on web pages. Users have become less patient and have less time on hand, while search engines still require users to dig through and extract information from the web pages to find what they’re looking for. In addition, users are spending more and more time on their mobile phones and other connected devices, which require a completely different kind of user experience for search.
When we talk about Search, keep in mind that Search, Discovery, Recommendations, and Serendipity are all essentially the same thing. Why? Well, to start with, one would need a comprehensive index of content for each of these things to work. This gives you a world view, so to speak. How that index is created has changed over time, and what goes into that index has changed. About ten years ago, the index only consisted of HTML pages, but that information has been changing. How the index was created was heavily focused on signals provided by HTML pages, links, consumption, etc.
Today, many social signals are consumed, including how often and how quickly an entity or URL is being embedded elsewhere, whether it is with positive or negative intent and sentiment, and is it trending up or down since last week/month. Search engines have mostly focused on the backend and infrastructure, and rightly so, because search requires a delicate balance between some of the most complex technologies, and a vast amount of infrastructure. Solving today’s user needs requires a different focus: a special blend of science, a finely tuned user experience, cutting-edge design skills, and a slightly different mix of engineering and infrastructure.
The question now is—how do search engines respond to this new world?
The answer, to put it simply, is to re-imagine search. The new landscape for search will likely focus on getting the answers the user needs without requiring the user to interact with a page of traditional blue links. In fact, there may be cases where there are no blue links on a search results page at all.
Search engines will keep assimilating content from many different sources and aim to provide immediate and rich answers. You ask a question and you get answers, nothing else. The user may not even type the full question. Search engines will have to become more and more personal, understand the individual user’s preferences, location, type of content preferred, context from previous search and browse behavior, signals from social graphs, and much more.
Search has been a pull mechanism for information and content, while social sites such as Facebook and Twitter are push. For search to succeed in today’s world, it has to become more push, which is why we at Yahoo! have been so focused on what we call contextual searches. A contextual search is when a user happens to be away from a search box, maybe reading an article on Yahoo! News, and comes across a name, or place that he/she wants more information on, yet they don’t want to spoil the reading experience and leave the page, open a new tab, and do a search.
With Infinite Browse, Yahoo! currently enables users to highlight the term and get a small pop-up search result out of that action, without leaving the page. Yahoo! also identifies and underlines interesting terms/entities on the page, so when the user hovers over the word or words, additional information is provided.
Imagine a future where this information is entirely pushed to you without prompting the search, so engagement with the content you want is immediately at your fingertips. This will prompt more and more searches to happen away from traditional search results pages, and will happen more in context of wherever the user may be—reading a news article and wanting to know more about a topic or entity, accessing information on a commuter train, getting recommendations pushed while writing an email or social conversation on that topic, and much more.
In the near term, innovation in search will provide more in-depth answers. For example, if someone types the name of a Major League Baseball team, they get a search results page with the team’s homepage and likely a couple pieces of recent news. In the next phase of search, you will type the name of that baseball team and without hitting the search button or leaving the search box, you will be presented with an interactive display that includes a link to their homepage, recent news, the results and box score of their last game, their overall record and standing in their division, a schedule of upcoming games, photos, videos, and social media streams.
How about searching for a restaurant? In search today, you find links to the restaurant’s homepage, address, phone number, and rating. In new iterations of search, you will type the name of that restaurant and be provided with its address and map, a view of its menu, the option to reserve then and there via OpenTable, see its ranking on Yelp, CitySearch, Zagat—along with photos, tweets, what your friends have said about it in your private social networks, and a quick and simple way to compare it with other similar restaurants.
The next chapter of search is going to be about providing answers and not just answers from Q&A sites (although Yahoo! Answers hit a billion Q&A last year). We obviously believe in these types of “answers” and leverage it heavily, yet there are plenty of other types of real-time answers.
Most search indexes are in the 10s of billions of URLs, trending towards 100s of billions of URLs. Information is dynamic and changes frequently. For example, the movies running in a theater next to you are changing every week, and the timings may change even more frequently. The San Francisco Giants score changes frequently too, as do the players stats. So, while Q&A sites are really interesting in solving a certain set of needs for users, they are only a piece of the puzzle.
But the rise of Q&A sites across the Web speaks to the underlying need for better answers. A new era in search is just around the corner that will make it easier to access the information, services and answers people are looking for. A list of links just doesn’t cut it anymore.