MODERATOR: Come on in. Grab a seat. We are going to get started. It is the morning keynote of the second night. And I’m dragging a bit, and it’s Yusuf’s fault because he threw a really good party for us all from Bing. So, once again thank you very much for that.

YUSUF MEHDI: Absolutely.

MODERATOR: I’m joined here today with Yusuf Mehdi. He’s senior vice president for Online Audience Business of Microsoft. And your job is revenue and market share, so it’s making sure people know about the services, and not just Bing, but all the online services, correct?

YUSUF MEHDI: Yeah, that’s right. Yeah, within Microsoft, we have a division that’s called the Online Services Division.

(Break for direction/mic issues.)

YUSUF MEHDI: So, yeah, my name is Yusuf Mehdi. So, at Microsoft, there’s a division for online services, and my job there is to run basically the business —

(Break for direction/mic issues.)

YUSUF MEHDI: All right, take it from the top. My name is Yusuf Mehdi, and I work at Microsoft, and I work in a group that’s the Online Services Division, which includes Bing and MSN in particular. The job I have is to — a couple things. One is to help gather the consumer and market requirements to build the products, to run the business of it, which includes the marketing P&L, the strategic partnerships, and helps on the strategy.

MODERATOR: So, it’s been a year. We’ve come a full year since Bing launched, and I think it’s been one of the most remarkable marketing onslaughts that I’ve ever seen. I really liked all the stuff that you’ve done so much that we talked about doing a little movie. So, we’re going to have some movie now, I hope, as a review. Cue the movie. (Laughter.) Oh, here we go, excellent.

(Video segment.)

MODERATOR: So, I mean, it’s just amazing because you had everything from the TV commercials to the product placement, and, in fact, this week, we had Stephen Colbert, I guess you have to pay him $150,000 now because every time he said Bing, you’re paying him $2,500 for his charity fund.

YUSUF MEHDI: Yeah, that’s right, we did a promotion with Stephen Colbert. He took full creative license on it, and we said — (laughter) — we will — you guys, if you haven’t seen it, you can go find it on the Web, you can Bing it. He said that he wanted to do a great thing for charity, to donate money to help the people in the Gulf. He said — we sort of did a promotion where we’d donate money every time he said Bing, we’d donate $2,500. And he took full advantage.

MODERATOR: He’s going right at it.

YUSUF MEHDI: He did a great job with it.

MODERATOR: I wanted to say, by the way, I mean, it was kind of crass really him charging you $2,500 a thought. I’m going to not give you a — I’m just going to give you a flat fee.

YUSUF MEHDI: All right.

MODERATOR: And if you want to make 150,000 in $1 donation, in addition to his, then I think that would pay off for you because, first of all, he’ll be so angry that somebody he doesn’t even know has gotten more money than he did, that he’ll mention you again. So, it’s like double dipping, but —

YUSUF MEHDI: I like it, I like it.

MODERATOR: — I won’t hold you to it.

YUSUF MEHDI: All right.

MODERATOR: But I’m just saying I think it would work out well — perhaps for me, but, you know, maybe not for your paycheck.

So, is it paying off? How do you know if it’s working? You survey everybody on the street and —

YUSUF MEHDI: Well, so I’d say, first of all, a couple things. First of all, thanks for being here. It’s a real honor and pleasure to be here in front of all these folks. Yeah, it’s been one year, and we’ve been working like crazy, and we have a long ways to go. And believe me, everybody on the Bing team at Microsoft knows that there’s a lot of work to do to improve the product to get to where we want to be, but for the first year, I’d say it was a good start.

Clearly, the answer to your question, Danny, is the way you know it’s working is if people use your product. You know, I know like a year ago we had a dinner, and we were, like, I was showing you the stuff, and Danny was going deep on — I was showing him a little bit of Bing before we launched it, and Danny, like only Danny can do, started going, really let me probe on this, let me probe on that, let me talk about this feature, and suddenly, I was driving home thinking, “God, I don’t know if this is going to really work. I don’t know how we’re going to do our first year, it’s a 1.0 product.”

But the good news is we’ve actually grown users, and we’ve grown 20 million new users in a year. That’s led to four points of market share growth. We’ve grown every month since we’ve launched.

So, I think that’s probably the truest proof case about whether it’s working or not is, are people coming and trying it and are they happy with it. Thus far, knock on wood, long ways to go, it’s working.

MODERATOR: Some of the ideas have been a bit out there. The Bingathon was there, and that seemed kind of crazy.

YUSUF MEHDI: Yeah.

MODERATOR: What were you thinking?

YUSUF MEHDI: Well, you know, I think is when you’re the number two you have to challenge, right, you have to get outside the box. I mean, like the little video said — and when we launched — our whole theme was no one is sitting around waiting for another search engine. No one is sitting around saying, “Oh, if I only had another search engine, life would be good today.” Because people were really — people are — very happy with what they had. So, we had to kind of get outside the box.

So, one of the things we would try to do is, we said, “Hey, when we show people the product, they really like it, but it’s how do you show it to them. So, we thought about how do we get creative, take some risks, and we thought, “You know, infomercial, no one is going to want to sit through an infomercial.” Hulu turned out to be a place where a lot of the target we were going after was there.

And we had some creatives, people who said, “Hey, what about this crazy idea of doing this thing called a telethon where we’ll have people use Bing and we’ll show features,” and we did that. You guys maybe saw a little clip of it. It was clearly a little over the top. I’m still wondering how it all worked out. But, yeah, we tried some stuff. We pushed the envelope.

MODERATOR: What seemed to have worked the best? I mean, “Gossip Girl” saying “Bing it,” is that paying off more than the regulator standard television ads?

YUSUF MEHDI: Yeah, you know, I think it’s a lot of things, actually, are working pretty well. I think the most important thing is definitely about being in the shows, and some of the shows being integrated, that drives the buzz where people talk about that. So, where we’re able to have a real, sort of, stuff in-show integration, and people said, “Hey, did you see that, I didn’t know we could do that,” that ends up being the most successful part of some of the marketing.

MODERATOR: And what’s been, sort of, not gone the way you expected, that should have done better, and we didn’t —

YUSUF MEHDI: Well, there are a lot of things that — a lot of things that haven’t — you know, should have gone — and I’d say, you know, at the beginning — this is also a true story. So, I was having a good day, we had a lot of good progress, and I got home, and in my home, the Mehdi home, Bing is clearly a very important thing to be using. And I don’t tell my kids, “Hey, you’ve got to do the following,” but they’re loyalists. And I got back, and my daughter gave me this thing, and I thought I’d show it to you because I carry this around. I’ve got this framed up in my office. She came back and she — this is a list of people who use Bing, and then the other engine. (Laughter.) This is Bing right here, there’s four names, Flora. There’s one boy. We call that boy. I know he’s got a crush on my daughter. That’s why he’s on this list. (Laughter.) So, I can’t even really count him. And then there’s a long list over here, the top two of which are her best friends.

And she comes to me and she says — and my girl is in for everything — but she comes to me and she goes, “Daddy, we have a lot of work to do.” (Laughter, applause.) I didn’t even ask her to do it, she just on her own.

So, yeah, in terms of things that, like, haven’t worked, this part here is, basically, there is a habit of using what’s comfortable. You’ve talked about the habit in a certain way. I think showing people, “Hey, we have this great new engine, we have a point of view on how we can help you make decisions.” People still said, “Well, yeah, but I’m comfortable with what I have.” So, we still have work to do to sort of I’d say push the envelope more.

We’ve done a lot of things that have — you know, we’re a kind of trying-and-try-fail-try-again kind of company. We tried some things like with Cashback, which is a program that we recently said we’re going to end in July. It was successful in a lot of ways but didn’t work out in terms of what we wanted to do — didn’t actually show the growth and change in behavior.

We’ve done a lot of experimentation in the product on the user experience. One of the things we’ve done is we bet very big with the left rail. You all know that’s super valuable real estate.

Some things have worked really well, but one or two things are not working that well, and we’re going to do some more revs, and you’ll see some more stuff in an upcoming set of releases where we’re going to try again, and take a big, big gamble, which only we could do, take those gambles, because I think we’re from number two, and those things will work.

We’ve done some distribution deals to try and get our product out in front of customers, and some haven’t worked out so well. The return hasn’t been worth the investment in money.

So, there are a lot of things that we have tried thus far that don’t work, and we have to tune and improve them.

MODERATOR: Now, talking distribution, you’re going to be on the iPhone 4. You’re also in the new version of Safari. Did you just have to write a big check to Steve Jobs for that, or just promise to use iPads, or how has that come about, and what’s —

YUSUF MEHDI: No, the good thing about the partnership with Apple is it’s been really a fantastic win-win, and it started first with our iPhone app. So, one of the things that has been kind of a natural fit for us is that first of all our vision for Bing is about helping you make decisions — it’s about empowering people with knowledge, which is a little different than, let’s say, index the world’s information.

The key to that is you have to understand user intent. The signal of what do people want to accomplish is critical to us. And the mobile phone is a unique platform for that because you know more about the user, because people don’t sort of share phones a lot — you know location, you know speed of movement, you know a number of things that are going on. So, that allowed us to do a better job building the Bing product on the mobile phone than on the PC, in fact.

The second thing is Bing has succeeded very well because of what we’ve bet on visuals. So, whether it’s our beautiful homepage or what we’ve done in the SERP, visual is what really has struck a core with consumers and where we’ve differentiated. Obviously on the iPhone you can take visual to a whole other level, because you have the touch interface.

So, we started out with a great app, and the app has been a huge success on the iPhone. If people don’t have it, you should definitely download it. And it’s been tops in their app store.

And they’ve seen that, and we had a lot of discussions. And obviously they said, “Hey, look, we’d like to — our customers are saying, ‘we’d love to try Bing,’ and choice is good for consumers.” And we, of course, are interested in getting out there. So, it was a very easy discussion.

One of the key things that we did is we did a lot of work on HTML5, which, as Apple announced recently, is a big focus for them. And we showed them some prototypes, which we demoed. They’re only prototypes that we demoed them at the worldwide developers conference at Apple. And those really lit people up there about all the things you could do with HTML5, and we’re going to bring that into market in the next number of months — we’ll bring those features out.

But that’s really the thing that I think struck Apple. They’re a company that’s about user experience first and foremost, and I think it’s a very nice compliment that they said, “Hey, we like what you’re doing, let’s work together.”

MODERATOR: But do they actually charge you for it? Do you have to pay to be included? Does everybody have to?

YUSUF MEHDI: Yeah, they have all their various terms, business terms with each of the companies, and we agreed not to disclose our terms. So, I can’t talk about those.

MODERATOR: OK. I’ll get Google to disclose their terms, and then you guys can disclose it.

YUSUF MEHDI: That will be fine.

MODERATOR: We can all disclose like whatever.

YUSUF MEHDI: That would be fine. (Laughter.)

MODERATOR: How about — what have you spent to date? I mean, I heard numbers, it was going to be 80 million. Did you spend it all, and are you out or —

YUSUF MEHDI: Yeah, why don’t you add some money?

MODERATOR: — and how long can you keep going?

YUSUF MEHDI: I’m calling Colbert back and saying, hey —

MODERATOR: Yeah, exactly.

YUSUF MEHDI: No, I mean, look, we’ve made a very big investment. I mean, we have certainly made a big investment to build a product. I mean, we’re — in fact, in many ways we’re investing ahead of the curve. We’re spending a lot of money on R&D in Microsoft Research to improve what we can do with technology. We’re trying to push the boundaries. We’re doing a lot of things in new areas where people have not done before, and also in distribution. As you know, we’ve spent a fair amount of money. I won’t disclose the actual figures, but, yeah. And you can see the TV ads, and anyone can value what we’re trying to do.

So, we’ve done a lot of pushing of ourselves to invest and to try and innovate in this space because we believe that at the end of the day there will be a great opportunity, not only for Microsoft and the company, but for the industry at large.

MODERATOR: But are we likely to continue to see the ads and Bing putting itself out there for the next year, two years?

YUSUF MEHDI: Yeah, I think you can. I mean, as we talked about earlier, we’ve actually had some great success because our share has grown, and people have come to see the product. And the more they see it, the more they’ve switched over and tried it. So, yeah, I think you can expect to see us continue to do hopefully some great innovative marketing over the next year.

MODERATOR: Until your daughter comes home, and she’s got more Bing people.

YUSUF MEHDI: Until we’re more balanced out and all that, that’s right.

MODERATOR: Let’s talk about some of the features. When you launched, you had a huge number of features. Many were already there at Live Search, but it was amazing to watch people sort of rediscover them when they were put out there under a new brand, and so on. What’s resonated and what hasn’t?

YUSUF MEHDI: Yeah, the — the things that have really resonated I think with Bing have been, number one, like I said, the visual orientation. So, the ability to harness the power of search and the power of the Internet by better visually understanding what’s going on was the number one thing I would say. So, the new SERP design that we had, the look and feel, it’s pretty amazing, and I think we got lucky, and we struck lightning in a bottle in the sense that people said, “When I use your experience it does feel a little bit easier to find what I’m looking for.” That was number one.

The home page in particular is quite good. It’s very hot with kids in schools where kids sort of say, “Hey, I want to start my day, I want to start by day with the home page because it’s a way to discover the world.” That struck a chord.

The third area is we focused a lot on helping accomplish tasks. That’s how we position ourselves as being unique and different is we’re helping you get tasks accomplished. And the bets we made in verticals like travel — so we had a bunch of great features, Price Predictor. If you’re going to buy an airline ticket, we have algorithms that will tell you if those fares are going to go up or down. That was a huge hit. A lot of stuff in shopping — we did a great shopping vertical. We did a local vertical. These all have had great success. They’ve over-indexed in terms of our results.

And then, even the little things — like, we did a thing with recipes that came out. And don’t tell me why, but that thing went nuts in terms of success. We have huge traffic. So, the little things like recipes, or we talk about snow reports or the instant answers in the main SERP, those are the things I would say — and then images of what we’ve done in images. Those are the areas where I think we’ve had real great success.

MODERATOR: Which of the verticals is the most popular outside of regular core search?

YUSUF MEHDI: I think — well, generally speaking, obviously images, video, very popular. In terms of what we have focused on and where our growth has disproportionately come from, travel, shopping, health and local have done really well.

MODERATOR: Then, in December, you rolled out apps for the maps, for Bing Maps, and so suddenly, we were able to do things, and you’ve added things, like, where the Foursquare check-ins are, where people were tweeting across from that. How has that been received?

YUSUF MEHDI: I’m glad you brought that up. So, mapping has also been really well-received. The map apps in particular, we’ve not done as good a job to promote them, I would say, but for the people who have gone to find them, it’s been a huge hit. People have really lit up.

If you guys haven’t seen it, if you go onto our maps product, maps.bing.com, there is, down at the bottom, a thing that says map apps. And you click it, and there’s a bunch of opportunity now where you can overlay visual data on top of the map.

So, the interesting thing is, you have this weird effort where you go and index stuff, and you separate geo location from a set of content, and then you try and reassemble that.

Well, what we’ve done is tried to take that to the next level and say, “Let’s reassemble all this great information and present it to you visually, so you can see the front newspaper of 100 newspapers around the world on a map, or you can see where people are checking in and becoming mayors in Foursquare, or you can see a number of other things that we’re doing on the mapping effort.” That’s been a huge hit. In fact, we have a new release coming out — I think it’s within a week — that allows you now to — developers to go and do mash-ins on their own. So, before we were building a lot of those, and now people can come in and build their own map apps. So, that’s been pretty successful.

MODERATOR: So, there’s an app for that map.

YUSUF MEHDI: There’s an app for that map, a map app.

MODERATOR: And the other thing, I mean, you had an amazing demo that you ran earlier this year where you’re doing the live video within the street-level photography that you’re doing. That’s live now, isn’t it, or that’s coming?

YUSUF MEHDI: We have it as a prototype up that you can see it. That feature is, I think, coming soon. We don’t have it live now; we have the live stream in. But, yeah, the idea is exactly that, Danny, which is you can take live data feed and then superimpose the geo location of it within the map.

One of the things that we’ve done for anyone who’s into it that I think, again, is unique relative to what anyone else has done is we’ve built a true 3-D image of what’s out there. Others have done sort of JPEG photos of it, and they’ve tried to build a 3-D model. We’ve built an actual 3-D model where you can, in fact, come in and layer things on, and then as you start to do streetside navigation, you can, in fact, touch the various buildings and move around. So, it’s very powerful in terms of what we’ve done in the maps.

MODERATOR: Now, you’ve got a whole bunch of new features that are about to come out. You’ve got something you’re going to announce today. We’ve got a demo for some of that.

YUSUF MEHDI: Yeah.

MODERATOR: So, let’s bring that up.

YUSUF MEHDI: Sure. Yeah, so one thing while that demo comes up, so we ship pretty regularly. We ship a couple times every quarter, and then twice a year we do a very — we do a big release. So, we have our big release that comes out in the summer that’s coming up, and there will be a number of things that we’ll do there, too, as I was saying earlier, push the envelope on visual, et cetera.

But one of the things that is coming as part of that release that I wanted to show you is one quick demo here, which is we have — one of the things that we have tried to be a leader on with Bing is harnessing the power of social and the social effort. And as you all know, a big part of the Web is static, but a lot of what is growing now is all of the information on Twitter, on Facebook and the like. And that information becomes a very powerful way to help tell you about the world or to make decisions.

So, some examples of scenarios I think of is — let’s say that Friday we’re going to go to a movie, I’m done with a bunch of work, and you say, “Hey, which movie should we go see? Should we go see Splice, should we go see Killers, two, I think, horror flicks?” And I don’t know. Well, why don’t we go up and take a look at what people are saying real-time socially about — hey, is this a good movie, is this not a good movie. That’s a better way to find something. So, that’s an example.

But what I want to show you here, this is a thing that’s in beta. If this demo goes well, we’ll tell the developers to put it live and you guys will have access to it in about 15 minutes. But let’s make sure the demo works first. (Laughter.)

OK, so what I want to do is, this is – first, let me start with — let me tell you what we’re looking at. This is the Bing.com/social vertical. This is a beta of it. And what it does is, this is where we pull together now the information, all that social information, and we’re adding a couple things.

First, let me start with the beginning, which is this is a new user experience for our social. On the left-hand side here, what we’ve got is we have trending topics. So, think of, this is like the tag cloud, what’s hot, what’s trending. And so you see World Cup, South Africa, Michael Jackson continues to be popular. And what you see now is we’ve added Facebook. So, we’ve now added Facebook and Twitter sources, and I’ll come back and talk about that in a second.

And then I can come up and pick some topics and see what’s going on. Here’s one that’s interesting, which is Strassburg. I’m not sure if anyone knows what Strassburg is or why it’s hot. But if you’re a baseball fan, you would know, in fact, that he is this new pitcher who just made his debut yesterday. I’ve been working so much I didn’t realize he’s actually on my fantasy team. My son came in last night and told me, “Hey, you’ve got this guy, I want him,” and I’m like, oh, I’d better find out what it is because usually I don’t get a good deal from my son. (Laughter.)

So, I come in and I see — what we do is we do a nice job here. We actually give you a summary of who is Strassburg; so just like we do on search mainline results, a quick snippet of it.

And then we have top shared links and public tweets.

But let’s go ahead and dive into Strassburg and take a look at some more information.

So, here are now all the real-time public updates coming from both Twitter and Facebook, and we see the shared links. And if I want to dive in, let’s go in and we take a look at shared links. This is a pretty powerful thing now. What we’ve done is we’ve taken the links that get shared on Facebook that are popular, and we don’t attach users’ names to it. It’s just the most popular links. And you can come in now and you can see what are the links that are being shared in Facebook or in Twitter. You can see here, here’s a public Facebook one that comes through. And I can scroll through, and I can see what’s happening here.

So, for example, what’s nice is, here’s a baseball video highlight. This is a relatively unknown guy. He came up through the minor leagues. And you can see here’s a fan house page where people really know about him, MLB.com, news, videos; so a nice way to get a bunch of information.

I can also then take a look and come and see the fire hose. So, this is all of the updates that are coming. These are either tweets, all of the tweets or all of the Facebook updates that are on fan pages only. We don’t actually take your public status — even if it’s public, we don’t show those here. We just show the stuff that’s on fan pages.

And I can sort it. I can say just show me the tweets from Twitter, or I can say just show me the postings from the Facebook public fan pages, and you can come in and see here all of the information that’s happening, and this is in real time. And I can either see the most recent or I can switch to say best match, and so the best match will pull up things like, oh, here’s information from the Journal, the Wall Street Journal, from a bunch of other sources, sports talk, so it gives you sort of more information.

And then if I want, I can even come in and I can go back and see the overall summary information.

So, what you get is you get some pretty nice efforts. And let me see if there’s a — there’s a feature we have here that I want to see if it’s actually firing live right now.

What we’ll do is, for some of these links, I think it will come in, in a different topic. Let’s pick up one other topic I want to show you just to close it off here. So, let’s take obviously a popular topic, SMX. And so you can come in here, and we can see some of the shared links that are happening, and we see those going in real time live. And I wanted to show you — let’s see.

MODERATOR: Quick, everybody tweet SMX.

YUSUF MEHDI: Yeah, everyone tweet SMX.

Well, it isn’t there, but one of the things that we’ll show you that we’ll fire is the ability to see how many people have shared a link. So, for example, I was checking on this just before we went live, and there was links where it says, OK, 10 people have shared this link. So, it’s another way to get, for example, information about what’s really hot, what’s really trending.

So, the idea here — and this will go live in less than an hour. As our developer, (Paul Yu ?), is online, we’re going to tell him to go ahead and give it thumbs up.

You’re able to now harness the power of what’s going on in the social community to help you make better decisions. We have here through our partnership with Facebook and Twitter the best access to the social information, better than what’s offered in any other search engine. So, if you want to really figure out what’s going on in real time, we’re going to launch this shortly, and it will be pretty good. (Applause.)

MODERATOR: So, obviously, a lot of attention has gone into the whole social area. You know, some people wonder if it’s just getting onto the latest trend. I mean, are you finding so far with the experimentation that it really makes a difference, and people are turning and saying where can I search Twitter, where can I search for social?

YUSUF MEHDI: Yeah, I would say I think we are probably halfway into that, Danny. So, like, as I said, there are definitely clear use cases where people really do see the power of social. And within the vertical, there are good examples. I think you want to get in on the main page, right? I think firing it in the main SERP is going to be critical. And that’s — and doing that well is important. I think we’ve tried it, some of our competitors have tried it; no one has done a super good job. We’re going to start now trending pretty well with this effort, and I think we’ve found the magic that will make that happen.

But clearly, intellectually, it makes sense that you would have the power of social to help add and enhance some information to the SERP, and I think that’s coming along.

MODERATOR: You had mentioned earlier the task orientation of things that are going on, and we’ve talked before, and I think one of the things you get most animated and excited about is when you talk about these search funnels or these search paths that people go. So, I wanted you to talk a bit more about some of that research you’ve looked at, and we’ve got some slides to actually show the audience some of these paths. Because I found it fascinating, and —

YUSUF MEHDI: Absolutely.

MODERATOR: — you — whenever I’ve had meetings with Yusuf, he’d have these stacks of stats and everything, and I’d just be like, “Can I just take pictures of everything?” So, we do have some.

YUSUF MEHDI: Yeah, so we have some slides I want to show you. This is — yeah, you’re right, this is sort of — this is my version of the geek-out, which is we get into all of this session history data.

Let me tell you a little bit about what it is before we jump into it. So, when we were building Bing a year ago — and this is what Danny is referring to — we said, “How do we go help people with problems that they don’t have, that they don’t have being met today,” as opposed to “Let’s just go 5 percent better in today’s search.”

And what we did is, we said, “Let’s really study what happens in search.” So, today, for most people now, you think, “Oh, search just happens like this: I sit down, I type in 2.8 keywords, and in milliseconds, I get an answer and I’m done,” and that’s a search scenario.

It turns out, it’s not. We do a ton of data mining about query logs. This is all anonymous, and then we have to ferret it out. So, this is a bunch of work that came out that we figured out. But basically, you take an anonymous ID, you know, ID6975, whatever, and we look at all the queries that came from that ID. People skip in and out about what they’re searching, but we ferret out what happens.

So, here are two examples. One is someone types in at 3:04 “family reunion Lake Tahoe” and they search. And they spend two minutes on this site, www.eventective.com. Then they go to a large Lake Tahoe house again. Then they try “retreat Lake Tahoe.” Then they spend four minutes on a bunch of different sites there. One of those is a vacation rental site. Then they go “retreat for rent Lake Tahoe,” 11 minutes. Then they go “large vacation home.” Then they spend two hours browsing a bunch of stuff about Tahoe vacation rentals. There’s a break in there. They come back again a little later, but again, “large vacation home to rent.” Then they have “rental that sleeps 30,” then “Lake Tahoe rental for large group.”

And what happens is after four hours in duration, they’ve gone to 27 sites, and they’ve typed 13 queries, six of which are the identical same query stream.

And you look at that and you say, “Is search perfect today?” You know, it’s not. There’s a lot of problems.

And you realize, “OK, how can we make this better for people? And we mine these things, as Danny was saying, and we have so many examples. There are examples then of, like, the classic shopping example — someone does a search; how often before someone buys? Take a guess. Let me ask you this way. What percent of people do you think from the first time they enter what we detect is a commerce query a purchase occurs two weeks or later? What percent would be two weeks or later? Anyone? Ten percent? It turns out over 40 percent of those purchases happen two weeks or later.

So, think about that. Someone types in a query, they’re not buying in that moment necessarily. In fact, 40 percent of the time they’re not going to buy for a couple weeks.

So, what is the right answer for that?

QUESTION: (Off mic).

YUSUF MEHDI: And then I’ll show you another — (laughter) — I’ll show you another one, which is, here’s one for mobile sessions because we were talking about mobile.

So, I was trying to spend some time. So, here’s a traditional search on mobile, OK, and it starts out “killing a tree stump,” “rotovator,” “landscape gardening”; week two, “top soil,” “lawn fertilizer”; week three, “what is grass, rye grass”; week four, “fungus mushrooms.” Then there’s a lawnmower query — “how to mow a lawn,” and then “garden.” So, you can see over seven weeks how people’s query chains have changed on the PC.

And you say, all right, well, how should we build our product because we know how people are moving through the service, right? We know what they might be looking for. That’s been interesting.

Then compare it with mobile, OK, and mobile is basically, first week, it’s “where to rent a rotovator.” Let’s type it again because they didn’t get what they wanted, “where to rent a rotovator.” Say that real fast 10 times. “Where to rent garden tools,” then a cultivator, and then they’re done. Basically, what you find is on the mobile phone — you know the same query I told you about, mobile phone — most of the queries on mobile phones are commerce-related. Well, what do you think is the extent of time before someone takes action on a mobile phone? We’ve talked about it at weeks, right? We know it’s faster. Any sense? I’d say the rough numbers, 65 percent of it happens in one hour — one hour and done. This is not a rich phone; these are not smartphones. Smartphones have more powerful Web browsing. I think that will change, but this is on some of more simple feature phones.

But again, the data will show you the way, right? The whole point about this is that when we went to go build a decision engine, we really wanted to understand what customers were doing to figure out how to change our product and build our product. And as Danny said, this is the thing that we just get so excited about at the Bing team. We’re just in this data all the time, trying to figure out how to help design the future of the Web in a better way.

MODERATOR: Now, the decision engine aspect of it, you know, it’s a great name, but are you getting there? It looks like you still have a long way to go to actually help people get to that final decision.

YUSUF MEHDI: Oh yeah, we have definitely, Danny, as we sort of set up for it, you know, with all the humility in the world, we have miles and miles to go before I think we’re doing a good job on decision engine.

We have some progress, obviously, in the sense that, you know, in the surveys we do with customers, they are saying that we actually rate well as a decision engine, but that it’s — and relative to competition, we rate pretty well. But that’s a long ways to go.

The work that has to happen is immense. We have to get much more of this type of data to understand. We have to get better signal data to really understand what people are doing. And then we have to build solutions to help accomplish that decision-making process, whether that’s task completion, commerce completion, what have you. So, we’re only in the very beginnings of that effort.

MODERATOR: You’ve got the deal with Yahoo! I think we get in place still by the end of the year, if I remember correctly.

YUSUF MEHDI: That’s what we’re shooting for. Yeah, so we’re shooting for in the United States, if things go well, that we could migrate the algorithmic and the paid search marketplace to one platform by the end of the year. That’s the goal. So far, we’re on track. But if we’re not ready, or we’re not ready to do it in quality, we will do it immediately after the holiday season, obviously.

MODERATOR: Now, that immediately gives you much more reach for both your own property and for the stuff that you’re powering. Do you move past them? Does that help get you past them? One of the goals would be to do that.

YUSUF MEHDI: No, the real — the partnership with Yahoo!, I think, is really kind of an industry-changing partnership around one basic thought, and this is what it’s about. It’s about creating an at-scale, number-two ad marketplace that can compete with the big guy. And by coming together, the hope is that for all advertisers, we make it simple and say, “Yeah, OK, now I definitely should buy two.”

Because before what happened is they’d say, “All right, I’ll buy on the leader, and then maybe I’ll buy Yahoo!, maybe I’ll buy Bing, maybe I’ll buy both, maybe I’ll buy none, because you’re both so distant.” Now, by being combined together and having approximately 30 percent of the ad marketplace in the United States, I think it becomes a no-brainer to buy there because I think the prices will be better and the return might be better. So, the key thing for the two of us is, let’s advance that ad marketplace and create more opportunity for the industry.

MODERATOR: You’re all now in — I call it the Bingplex. I don’t know if you guys have a name for it. But you’re in this big building now, in Bellevue, that I think has 26 stories, and I was amazed that there were that many people that were filling up the whole place. You’re off the main campus. How has that changed things for the team? Do you feel like you need to call into the mother ship, or it’s like, “Hey, we’re going to do whatever we want”?

YUSUF MEHDI: Yeah. No, no, I mean, it’s changing what’s — there’s definitely a great esprit de corps along the Bing team now because we feel like we’re all together. Before we were kind of diffused. Everyone is very close together. It’s more of a — it is more of an urban feel now. We’re in the city of downtown Bellevue there. So, there’s greater esprit de corps than there has ever been. I think part of that is, also, we just have some momentum, and that feels good, too.

But definitely very tied with Microsoft, I mean, in terms of we go to the main campus a lot for a lot of meetings. A bunch of the developers there from Microsoft Research are helping us do the innovation in the product. So, we definitely feel very tightly connected with Microsoft, but it’s a nice esprit de corps. There is a real buzz within the building of what’s going on, and it’s a great feeling.

MODERATOR: Have you guys gotten your logo up there yet? Is that going to come? Is it going to be up at the top of it?

YUSUF MEHDI: No, actually it turns out there’s a — I was actually pushing to get the Bing logo on the building, and there was — there are some issues with, I don’t know, I think the city or policy, or what have you, but as soon as I get free time, I’ll go back and figure that out.

But one of the things, I don’t know if you saw it in the video, one of the guys again, one of the creative guys on our side figured out how to turn the lights on at night. So, when the lights go out, it spells Bing up on this building, on a couple hundred-foot building, and that’s pretty cool. So, we are — we find ways to overcome, and sometimes those become better. (Laughter.)

MODERATOR: We’ve got time for two, maybe three, questions. We’re going to just go, actually, to a microphone live. So, put your hand up.

YUSUF MEHDI: And one of the things while those are coming up, one thing I forgot to mention is we have — one of the other things that’s coming this spring is we have an update on Webmaster tools. That’s an area where we have been sorely behind, and no one would know that more than the folks in this room.

Nonetheless, we’ve made some big progress, and I think at 10:30 we’re doing an update upstairs with Eric Gillmor and I, and I think and Stefan Betts (ph) are going to show it upstairs, we’re going to show you our new webmaster tools that have just come out, and I think they’ll be a step, a small step but a step in the right direction, and it will be great for folks in this audience.

MODERATOR: And we’ve got a write-up at Search Engine Land as well where you can get information on that, too.

YUSUF MEHDI: Awesome.

MODERATOR: OK, I’m going to leave it to you, Michelle.

QUESTION: Hi there. I was hoping you could speak to the roadmap of when Bing will be powering search for Yahoo!, and also, with respect to the display, rich snippets, game support, everything that Yahoo! has had, whether Bing is going to be supporting that or be giving Yahoo! the autonomy for it so that we can plan accordingly our strategy with rich media, rich snippets and how that’s all going to work.

YUSUF MEHDI: Sure. Yeah, so on the partnership with Yahoo!, as I was mentioning earlier, the plan right now is to try and migrate our algorithmic engine over, and our paid search marketplace before the end of the holiday in the United States. And obviously, we don’t want to get in there right during commerce shopping season. So, if we can make it — and I think we can right now but we’ll see — if we can make it, it will be before that. So, late fall going into the winter would be the migration time period, and that’s when we’d flip over.

The way it will work is we will power the algorithmic results and the paid search results. Yahoo! still has the flexibility to add things on top of the results, and so they’re able to do innovation and add answers. So, we make available to them everything that’s in our search API, including sometimes we’ll have snippets, and they can choose to use those or they can choose to add theirs on top. So, they could probably update you more what they’ll do.

But certainly everything that we provide for our own Bing will be available in there, so there will be one place where you can showcase any work you do on both platforms.

QUESTION: The four major metrics companies — Nielsen, Compete, comScore, Hitwise — all pretty much estimate what’s going on with the search engines. Search engines traditionally don’t really publish internal data.

Would Bing take a brave step and publish at least some sort of summary statistics so that people can really see what’s going on with the Bing marketplace?

YUSUF MEHDI: And is there something specific that you’re looking for that you don’t feel like you’re getting, just so I have it?

QUESTION: Well, I’m personally interested in numbers of users, user sessions, queries that produce click-throughs, basic — what to me is basic data that would help a search marketer analyze your demographic. We can go to places like Quantcast and Compete, and they estimate what’s going on, but those estimates vary radically. Nobody really has any clue as to what’s happening with the search engines. So, it’s very difficult to plan a marketing campaign based on search engine when you don’t have reliable data.

YUSUF MEHDI: Right, OK. I hear what you’re saying.

I think — let me follow up with you on some of those, because some of the things that — some of the things they provide, the aggregate data –comScore in particular we spend a lot of time — I feel is pretty much in line with what we see internally. So, I think that we actually track quite well with them on that, and I would feel pretty good about that.

If we publish public data, the problem that is you can’t really compare it against — across a number of engines. But I think some of the things you’re talking about I want to go back and double check about whether, in fact, some of that data is available. And if not, and it could be helpful, maybe there’s ways we could do that, and I’m open to doing that. So, let me follow back up on that, and then we’ll see where we are on it.

MODERATOR: We’ll do one last question, but also before I do that, if you’ve got a question you wanted to ask to the audience, I’ll give you at the end of that. But hands up would be helpful for them, so if you’ve got a yes/no type of thing. Yes?

QUESTION: So, there has been a lot of questions on the Yahoo! side about if Yahoo! will continue Yahoo! Boss, Yahoo! Site Explorer, but let me ask you from the Microsoft side, there is lots of advanced search operators that Yahoo! supports that Bing does not or has disabled. Yahoo! in their Yahoo! Site Explorer offers greater link numbers than Bing Webmaster Center.

So, are you going to do the numbers and the features that Yahoo! is currently providing, or are you going to go beyond that? What are your plans?

YUSUF MEHDI: Yeah, it’s a good question. That’s one of the things that I think we’re in real-time discussion right now with Yahoo! about how much of that functionality they provide today we want to take forward. So, we’re, actually, right as we speak, we’re in the middle of planning, OK, what do we do in that particular area, what features do we offer, do we want to offer some new things, some of the things in the past. So, we don’t have any firm decisions yet. Right now, Yahoo! engineers are joining Microsoft. So, we’re now collaborating and sharing ideas. So, just stay tuned. We’ll come back and give you some plans on that in the near future.

MODERATOR: They’re especially watching you to launch Bing Site Explorer, and you’ll make them all very happy if you do it.

YUSUF MEHDI: Got it, all right.

MODERATOR: Questions for the audience that you have? I know I put you on the spot. I didn’t tell you.

YUSUF MEHDI: Yeah, you put me on the spot. Let’s see. I don’t know if I’ll want to do what my daughter did and do the comparison on features.

I’d say — let me think, what would I want to know from this particular audience on Bing? I don’t know, do you have any ideas of what would be good?

MODERATOR: How many of you have been using it? I’ll ask and then kind of see there. So, you’ve used it. How many of you are using it on a regular basis? And I’m kind of curious, I mean, you guys study search, and you’re talking with people about it, as well. How many of you know people who have actually switched over? OK. So, I think that’s correlating well with your daughter’s survey. (Laughter.)

YUSUF MEHDI: If I can get everyone’s name, I just want to write it down. (Laughter.)

MODERATOR: There was a boy back there. (Laughter.)

Did you pay her like $75,000 for that market research?

YUSUF MEHDI: No, no. She’s the Mehdi family.

MODERATOR: How old is she?

YUSUF MEHDI: She’s 8 years old.

MODERATOR: OK. That’s good research. I think there’s a future —

YUSUF MEHDI: Second grade.

MODERATOR: Well, I started off by saying it’s been a year since you launched, and so we want to celebrate your birthday because we have a cake. So, if can bring the cake out, yes, that’s right. And yes, that’s right. And I’m going to make you all sing “Happy Birthday” to Bing.

YUSUF MEHDI: No. No, no, no. (Laughter.)

MODERATOR: But if you do, it will guilt him into making that further donation. (Applause.)

There we go. I’m going to do it: [singing] Happy Birthday to you.

You’re all in violation of copyright. DMCA notices will be issued shortly. (Laughter.) We’re going to cut this up into about 1,500 pieces. (Laughter.) No, actually — and Yusuf doesn’t know anything about this. We’ve been working with his team behind the scenes. There are like over a thousand or 2,000 Bing cupcakes for their birthday.

YUSUF MEHDI: Is that right?

MODERATOR: Yeah. (Laughter.)

YUSUF MEHDI: Well, actually —

MODERATOR: They’ll be waiting for you at the Bing booth, cleverly hidden back there.

YUSUF MEHDI: Smart. Good team there.

Seriously, thank you very much. I think — I mean, honestly I think this is — I don’t think we deserve cake and candles. We’ve got a lot of work to do. (Laughter.) Definitely. But I would say, I will share in the moment in the sense that I think, if I can, us as an industry, this has been a great year, and we really appreciate all of the help and support and engagement you guys have had because that’s what has allowed us to get a start. And hopefully, we can have another talk a year from now and still have some positive things to say and celebrate more good work.

Because I think we’re working in an amazing industry. I love the search engine industry. I wouldn’t work in any other place. It’s just too much fun, all that data wallowing, et cetera. It’s a great thing.

So, I think this is this — I’m glad the guys did some cupcakes. I didn’t know about that. I’m glad we did some cupcakes for everybody. It’s a celebration for all of us. So, thank you very much.

MODERATOR: Thank you, Yusuf. (Applause.)