If I had to sum up the work we’re doing at Bing in a sentence, I’d look to our vision: “Empowering people with knowledge by computationally understanding user intent.” Vision statements are often criticized for being lofty and meaningless, but at Bing, this vision really guides not just how we think about the product, but how we think about the future of search generally.
The Web has changed in the past 10 years, and more importantly, the way you use the Web has changed dramatically. Far from just finding pointers to interesting articles, the Web is now the transport that enables people to conduct tasks that were previously confined to the physical world in an efficient, near real-time virtual manner. Imagine all the things you can do because of this amazing Web transport: book a restaurant, hail a cab, conduct all your financial transactions, buy a vintage T-shirt, reconnect and stay connected to people you’ve lost touch with, tap into the cultural zeitgeist in real time – all these things are now possible because of the globally dispersed network that we think of simply as “the Web.”
Indeed you’ve changed from simply consuming information on the Web to using it as a vehicle to interact with information and services. The challenge? The majority of Web users start their day with a search. It recently has overtaken e-mail as the most used service on the Web. But the keyword model that we’ve all grown to love is beginning to show its age. Indeed, there are many cases where the keyword-to-URL mapping system pioneered by the early search engines still serves a great purpose, but as people increasingly move from simply finding information to conducting their everyday tasks online, the search paradigm needs to shift.
Task-Oriented User Experiences
At Bing, we’ve focused on pivoting the user experience around what someone is actually trying to do. We don’t believe the future is more links on a page, but in experiences that map to what you’re trying to get done in your real life. For example, you can see how Bing morphs itself when you look for U2’s latest concerts or when you want to make sure you’re packing the right clothes for the weekend getaway to Vancouver.
But we think we can do even more. Our visual search technology was developed in response to the fact that people, especially when they found something with which they were not familiar, weren’t able to easily construct a query. Imagine trying to find the best digital camera using only words. It might look like this: “SLR digital camera over 14 megapixels canon.” Even if you knew all the words you wanted to use, the engines wouldn’t do a great job understanding or responding to your request. With visual search, we’re able to guide a person from a very generic query (“digital camera”) to an experience that helps them narrow down their choices and ultimately gives them a tailored experience to make a better informed decision.
Partnerships Help Deliver on the Promise
To capture user intent, our engines need to be more agile. They must respond quickly to changing user needs and behaviors. The question we ask at Bing is how do we do a better job in responding to tasks where people require more than blue links? In many cases, we don’t have the data in the index that can be used to construct these vibrant responses to task-oriented questions. Part of our strategy is to partner with providers of rich, structured data that we can leverage as we build the dynamic experiences. We think we need to build Bing as a framework that can leverage the breadth of the online ecosystem to do something that standard Web crawling can’t provide.
Our most visible partnership in leveraging the online ecosystem is probably how we include Twitter content; this work really demonstrates our commitment to distilling knowledge from data. Rather than just inject tweets into core results, our Twitter search technology helps take out the noise from the stream and shows you the most important things people are talking about in Twitter. But you can also see the power of partnerships in things like our recipe search, which combines data from many recipe providers into an interface that helps you find that perfect chicken saltimbocca recipe without having to go to a dozen sites.
More Visual, More Interactive
The last thing people tell me they love about Bing is how it feels more like a beautiful application rather than a text-based DOS application. We’re very aware that visual experiences can sometimes hamper performance, but in places where making Bing more expressive can help you get something done faster and with more confidence, it just makes sense to leverage peoples’ innate ability to process things visually. Our work in Bing Maps is a great example of this – being able to re-associate data on the Web (like tweets, local blog posts, or user-generated imagery) with its originating place in the real world is a more natural and effective way to connect your virtual life with your real one.
The Road Ahead
It’s amazing how far search technology has come in a decade, but as we’ve said in the past, the promise of search extends far beyond our current reach. As search marketers and search professionals, you’re an integral part of driving people to expect more from their online discovery experience, and we’re honored to be a part of the journey.
This column originally appeared in the March 2010 edition of SES Magazine.
Time is running out to feature your company in our inaugural Mobile Vendor Reader Survey.
Marketers create personas to better understand their target audience and what it looks like. If marketers can understand potential buyer behaviors, and where they spend their time online, then content can be targeted more effectively.
What’s behind a successful data-driven marketing strategy?
Audience targeting can be challenging in social media, especially when brands make quick assumptions about their target users. How can you avoid generalisation and what are the real benefits of it?