In our last interview with Microsoft UK’s Head of Search, Dave Coplin reveals details about the company’s redesigned search engine, Bing, talks about its new venture, So.cl, and reflects on the fusion of social and search.
- It’s been a busy year on all fronts for Microsoft, including search and social. The overhaul of Bing, now combining search queries with the social graph, then the launch of So.cl – merging social networking with search. How will these new developments change the search and the social experience respectively?
- Let’s talk about Bing first and then we’ll pick up on the other social stuff. We think there’s a massive opportunity effectively to change the fundamental premise of how search engines work. And if you think over the years search has been based on this thing called relevancy, and it’s about giving you what we think is the most relevant information to your query. But relevancy is only part of the story. Let me give you a trivial example.
If I’m standing in the middle of Glasgow and I do a search on my phone for sushi, looking for a sushi restaurant, any search engine can give you the ten nearest sushi restaurants. And that’s just relevancy. But actually when you think about it, that’s not that useful. From these 10 restaurants, how do I know which one is cheapest, which is the best, which is the worst, which is the one that’ll be open tonight? I don’t have any of that information; I just have proximity to where I’m standing. What I want to know is if I can start to move from the world of relevancy into a world of trust, which is a warm human signal rather than a cold binary algorithm.
So what we’ve done by integrating Bing with Facebook (and it’s only available in the US, I have to be clear, but it will at some point come to the UK)… so now if I search for sushi on my mobile phone, I’m tapping into Facebook and I could see which of my friends have liked the restaurants that are on my list. So if I know that my buddy Bill has liked a restaurant… because I’ve got a relationship with Bill, I know who Bill is, we’re mates, and I know Bill’s taste in sushi is absolutely bloody awful, and I see he’s recommended a restaurant, I’m never going to go there.
So with this example you can see how we move away from relevancy into a world of trust. And I know, for example, that my other mate, Richard, has spent seven years in Japan and he’s recommended this place. Fantastic! I’m going there! I’ve moved away from this blind relevancy into a much warmer, human place where trust is really the main currency. Trust is really important. And that fundamentally changes the way search works.
There are times when this works, such as when you’re looking for a movie or buying a camera. But there are times when it doesn’t actually work. Just because all my friends like Kylie Minogue doesn’t necessarily mean I should like Kyle Minogue. So that’s why we have sectioned it a little bit so you could see the results, you could see how the social stuff overlays but you’re left to make your own decision. It’s up to you to decide whether the social stuff is important or not. We’ve built Bing that way because it’s a much clearer picture of the information.
And underneath all of this is a philosophy that basically says that people are no longer using search to find things, they are using search to do things. We think search is about doing stuff. It’s not about finding information, it’s about buying the camera, finding a restaurant and booking a table, finding a job – all of those things – and making them happen. So the social stuff is a crucial part of that strategy. The algorithm can only get you so far, the social graph is just so rich and warm with humanity that by blending the two together you kind of get the best of both worlds.
The other good thing that happens is that it brings to life the historical value of your social networking. For example, while we were doing some tests with this in the office, one of my colleagues was looking to buy a mountain bike and about 12 months ago one of her friends had ‘liked’ a mountain bike as she had been shopping for a bike herself. 12 months ago my colleague wasn’t looking to buy a mountain bike – she may have seen it pass through in her Facebook history but wouldn’t have paid any attention. But in a world where the social graph is connected to search, now that she’s looking for a bike it actually will show that: “Hey, listen, your friend – she liked this bike; she bought it.” And again, because you’ve got a relationship with that individual and you’ve got some trust, you could use that to influence your decision.
- Is your new project, So.cl, not essentially doing the same thing – blending social with search but in the form of a social network?
- Calling it a social network, I’m not entirely sure is correct. But what it is, as we found, is that in certain situations sharing your search streams, the thing you’re looking at, looking for, amongst a pool of your friends or colleagues, is actually quite valuable. So the examples we have are…think about all our researchers – we have 11 different labs around the world, working on the most incredible stuff that’s probably like five to 10 years away (Kinect came from Microsoft Research).
Typically they’re separated by geography – they’re all around the world working on similar projects. And we thought: “We’re going to have a bunch of people who will be really interested to know what their colleagues are looking for.” If I’m working with a friend who’s maybe based in India, and he’s working on something similar to me, I’d love to see the kind of articles he’s searching for, the things that he chooses and the things that he’s found because that’s going to be really helpful to me and my project. And today sharing that search stream is actually really difficult so what social does is actually make that process easy. It’s done through a Facebook log-in or a Windows Live log-in and it enables people to proactively share their search streams.
It’s not something people would want to do all the time but in those examples, it actually makes a whole load of sense. It’s come from our Office Labs and it really is an experiment to see if this stuff could work and whether it’s of value, and really, we’ll see how it goes from there. In many ways it’s an extension of previous searches; there’s another piece of research that we did that showed that actually people often repeat the same search.
- How likely are people to be willing to share their search activities?
- I think it’s a contextual thing. I think in the kind of scenario I talked about, people are really willing to share and they see great value. For example, I have a couple of counterparts working in different parts of Microsoft. I’d love to see the stuff they’re searching for and I’m sure they’ll be interested in the stuff I’m searching for. But for everyday consumers, I’m not sure they will get value and benefit from that service. There’s also a cultural aspect to it. People in North America, for example, are so used to sharing every aspect of their lives, they think differently about this from how we might do in Europe, for example. The concept of sharing your search stream is probably a bit bizarre to most people. In North America it’s slightly different. It’s really about testing that service and seeing what will happen. We know it’s valid for a large niche, if you like, of researchers, students, even people working in offices, and working on projects. For everyday consumers, probably it’s not something that they would want.
- You said the revamped Bing is yet to be rolled out in the UK. When can we expect that to happen?
- It’s going to happen in a number of different ways. Some of the features that came out in the US, some of the new interface, design, some of the new look and feel will come to the UK relatively soon. We’re still working on the timing for the social stuff. As you can imagine, there’s a bunch of work that’s taking place with Facebook to make sure everything is in place for that. That one is potentially further away.
About the expert
Since joining Microsoft in 2005, Dave Coplin has worked across a wide range of sectors and customers, providing strategic advice and guidance around the cost effective use of technology in relation to their business needs. As an established thought leader in the UK and having spent a considerable amount of time in the Public Sector providing leadership and guidance around key technology policy issues like Cloud Computing, Open Government, Open Data and the “consumerisation” of IT, Dave is currently working as Director of Search for Microsoft UK ’s Consumer and Online business, focusing the spotlight on the power and potential of search and the way it holds the key to society’s effective use of all that technology and the internet has to offer.
