It is one of those rare glorious days in Redmond, Washington with the sun out and not a spec of cloud in the sky. I find myself outside Studio F, one of the more than 100 buildings that are spread across the town that stitch together the Microsoft headquarter. Studio F is where the core Microsoft Windows Phone team operates from, headed by Joe Belfiore. The Microsoft comms team ushers me into a massive conference room, where I am told the core team comes together for brainstorming and discussing product milestones.
Over the next hour, Belfiore takes me through the evolution of Windows Phone since 2010, why he took some decisions that were necessary for the ecosystem and why he believes Windows Phone has the “app gap” and how he hopes Windows Phone 8.1 will address that problem. We also talk about Nokia and why having Nokia within Microsoft is important for Windows Phone.
The fascinating bit for me in this wide ranging interview for me was when Belfiore talks about addressing the vertical and horizontal dimensions of a platform and how Microsoft had been managing the vertical dimensions and how he is working on the horizontal ones. It gives some perspective on why Microsoft took some decisions in the past.
Of course, we talk about Cortana, WhatsApp and Moto E. Read on for the edited excerpts of the interview.
I was there when you introduced WP7, the queues were insane and I was probably one of the last people to get a seat. But you know, in 2010, the way you introduced WP7 and what was typically happening in the smartphone market, it was certainly something very different and your whole presentation was really exciting. I think people were actually surprised you changed it so much. It was nothing like anything that was present at that time. No icons as such, it was very fluid and sort of very interactive as well without being very intrusive. So what happened during that time — between WP 7 and WP 8.1?
Here’s the way I talk about that. I think what you said about WP7 sort of goes to the heart of what we are trying to do. We started working on WP7 and wanted to have a unique point of view on what a smartphone was all about. We picked the area of being personal and we are essentially trying to build a smartphone for each of us and not all of us. So Live Tiles was the manifestation of the idea that your homescreen would show you the data, people and information that you really care about you. The biggest symbol was to put your group of people on your homescreen to see their social network posts or photos.
So we knew that we had a unique take on that proposition and in the beginning we were changing from a model where Microsoft did not have a strong point of view of what the customer promise was and we wanted to change that. We faced this trade off which was that we had to create a product using two dimensions.
There is a vertical dimension of what is any one user’s one terrific experience and in that vertical dimension we needed a lot of software that we didn’t already have. We didn’t have a phone dialler, so we had to write one for ourselves. We didn’t have an app platform or a music experience that was full and complete and would allow you to buy tracks and sync them to your PCs. We didn’t have a great camera either and we have to do all this ourselves. We needed the value proposition of Office, Xbox Live and so what we did was start off by focussing on the vertical experiences and not focussing on horizontal breadth.
So what happened was WP7 launched and generally, my perspective was what people felt that this was interesting and unique. What they thought was if you live in one of these four countries and are on this particular network operator and are satisfied with this particular mid-range CPU, then it’s great. But the problem was these were not very many people. So since shipping WP7, mostly we have been “eating our vegetables” to get our value propositions to a much wider audience and there is a whole lot of dimensions for that.
In WP7.5 we filled out some feature gaps, went to some more countries and started to diversify our price points. That is also when Nokia came on. But even then we didn’t have broad price points. With WP8, we made a big technology change to get to the shared kernels between Windows and Windows Phone, which put us in a position to achieve the breadth. This is because the Windows kernel has a bunch of device drivers for a bunch of parts that OEMs want to put in a wider range of phones. This allowed us to get to a more flexible operating system foundation that we could broaden.
The next three updates that we did, each one of them in some meaningful broadened the way people had access to Windows Phone. So we went into China for the first time, we got the Lumia 520 done so we could target a low price point that we hadn’t been to. We also had the Lumia 1020, which was a high-end flagship value proposition which we hadn’t been to before. I think between WP 7 and WP 8.1, we have essentially tried to broaden and get the infrastructure you need to sell to the world.
Now, with WP8.1, some parts like dual-SIM is a huge thing and we can finally reach out to them. You see some of the Nokia phones that have been announced, and we will continue to broaden that price range and now we are also adding features like Cortana. On the one hand, we are closing feature gaps that honestly aren’t that huge, and we are coming up with our own unique point of view on that. So for example the Live Tiles, your phone needs a homescreen but we wanted to do something that matched our brand and was aligned with our value proposition. Cortana is again something similar. It is the first truly personal digital assistant that knows you and you can have a relationship with it. It is sort of the best of Siri and Google Now, with our point of view in it. That is how I would describe what we are doing here.
One of the things that become very clear after WP8.1 is that this is something you and your team were working on between WP7 and WP8.1. It was like a transition phase to enable this, but at the same time Android has moved really fast and there is a sense that WP as such remained stagnant despite all the new things?
I think the reason for that is it depends upon each person’s perspective. So if you were a customer in the US who bought a Lumia 900 in the fall of 2012, as an individual customer in the US with your American Lumia 900, you didn’t get many features on your Windows Phone. But you did get Instagram, which we got done by working with those people and you also got a bunch of other apps you didn’t have before. The reason that happened was because we were doing a lot of work that got lots of other phones sold, which increased the install base and made the app vendors more likely to target the phone.
I’m sympathetic that an individual end user might have felt like they didn’t get new operating system features from us, but the work we were doing was necessary to make the ecosystem healthy and you benefited from that by getting more apps.
But then at the same time there were visible glaring gaps like iOS copied the notification center long back and we get it now, but people have been talking about it for almost two years now. And people ask why does WP not give me such a notification center. So what’s the reason behind not getting those features?
The reason is that we made a decision that the more valuable thing to do broadly for the health of the ecosystem and for any individual customer was to grow the ecosystem size. This meant OEMs would make a wider range of devices at a wider range of price points and software developers would write the apps. The biggest complaint from users though has been that they wanted apps. So we were much more focussed on the infrastructure that you need to have a large ecosystem so that the apps show up.
You could say that we could have written more code, and in truth we wrote a lot of code to get lower cost phones distributed to mobile operators around the world and to get an app platform that you could target both on the phones and the PC. I think that was the right trade off. It’s not like I don’t think the American buyer of the Lumia 900 didn’t get his Action Center until now, but I think our decision to focus on the ecosystem for more apps was the better choice.
With WP8, people who had just a bought WP7 device suddenly realize that they wouldn’t get any more updates. Was that a very tough decision?
It was a very tough decision. We had many difficult and emotionally charged discussions internally and the thing about such large disruptive technology changes is that the sooner you make them, the longer you get the benefit and fewer people are negatively impacted. If we had waited to do that, the install base of people who wouldn’t get the update would have been bigger. So we tried doing it as soon as we reasonably could, once the technology was in place for the Windows kernel to be ported to ARM. It is unfortunate for those customers who got stuck like that, but it is good that a lot of app developers have continued to target them. But we think the overall better thing to do is to ensure that there are great phones at wider range of price points that have a common platform and get the same apps and we are already two years from the day we took that decision.
Let’s talk about WP8.1 can you take me behind the scenes of when you started working on 8.1? How did you decide about the feature list, and also what’s your take on the gaps and doing them differently?
The process for us probably started in the summer of 2012 before WP8 was done. The way it was done is that a set of people with different set of skills and backgrounds start thinking similar to how we think about cooking a meal. So what are the things this meal should have, like it needs some proteins and vegetables cause it is better to have some things that are healthy and it should have some desserts and so on.
So I would say at the time we knew that dual-SIM was big functional and market access gap, so I characterise that as a vegetable. It is delicious, but it is not a dessert. It is not something radically different from what the competition has, although we try to do our unique thing on it. So we characterized that it was a thing important for the business and the health of the ecosystem.
The biggest one of those was focussing on the app problem. We asked ourselves what’s the thing that is most impactful to making the ecosystem and the answer was getting developers writing more apps. So probably the biggest technical investment that we made in WP8.1 was the unified platform between the phone and PC. We knew that there was a class of people who wanted their apps to be on all platforms so they will come, some of them are open to a new platform but want the economics to work, so if you work that out they will come. So we worked through these things and the meaningful apps that were left were people who didn’t have enough resources or were having such great success on iOS and Android that they just wanted to keep investing in that, rather than investing on a new platform. The thing that we heard from this remaining set of app developers who hadn’t committed to us was that we needed more volume. So by doing the technology work that targets the phone and the PC together, we have a much larger volume to offer the app developers that makes the app ecosystem healthy. So that was the big choice similar to a protein or a vegetable.
And then there is the dessert. We spent time with our design team and the marketing team and there were two different documents that we write to characterize what we want to build in the release. One is the ‘Storybook’, which has the marketing stories about what we want to highlight for end users. The other is the ‘Planning Doc’, which has the priorities for the release. So the first phase is making all these decisions. So we decide on the proteins and the vegetables, we decide on the dessert — in this case Cortana — and in all these cases we used to think what would be the volume and what would the end users like and what technology assets we have in place to make it plausible and excellent. We knew Bing had evolved to a point where it had good data, where the speech technology was becoming excellent, so we could see that it was a good time to focus on such a scenario.
So you look at this recipe and then we communicate it to our team and say that these are the priorities. So we have to have an app platform that solves the app gap, we want scale to reach more users like opting for dual-SIM. We then roll them out to the team, there are checkpoints and milestones and everyone works on them. In this room, after every milestone we crowd over here and each sub-team says here’s what I’m building and all the leaders of these teams are here to look at each other’s work.
You spoke about covering price points, right from the Lumia 520 and going up to Lumia 1520. But what we have seen in terms of numbers it is your 520s, the 620s are doing extremely well. But as soon as you go above $400 (unsubsidized), that is where people prefer Android or iOS. You have excellent hardware, best possible camera that you can fit in a phone, great design, great battery life and stuff like that. So what is it that is not clicking?
That is a good question. I think that there are a couple of things. For one, it is easy to forget how new Windows Phone really is. If you work in the industry and you track these things, then you knew about it way before we even launched it. For most markets, it hasn’t been there for very long, may be just over a year. So one fact is that people are just getting to have awareness. What’s happened is that the products at the low-end have a very strong value proposition – iOS is not present there and although Android is present, I think our experience is better. We are more smooth, we are more reliable, all the apps run, and in the markets like the US where phones are all subsidized, consumers don’t draw that distinction. In markets where people are paying for their phones, that is where our low-range phones are doing well.
In the high-end segment, you asked what is missing. In this segment, people have a broader expectation of apps and they are less willing to forgive the small number of apps that we are still missing. So we are not at the 100 percent of apps that Android has, and if you are buying a high-end phone, you are more likely a repeat buyer who has already having one before and you are already used to that ecosystem. And you are more likely to be a sophisticated user who understands the breadth of apps and we are still at a disadvantage there. I hope we don’t have that disadvantage for that much longer, but today we still are. So I think the lack of apps hurts.
The other thing that has been a factor is that the Android ecosystem being an open sourced platform, the OEMs use it to do all kinds of ‘fast to market and experimental stuff’. So what happens is the very latest spec shows up on Android first, because the OEMs are all taking the source code and building as fast as they can. They won’t all be first, but somebody will be and the technology that they use to get there first with the highest spec was Android and the flagships phone tend to launch that way. It is the first phone to have the latest Qualcomm processor, first device to have 2GB RAM, etc.
Our model is not setup well to play the “bleeding edge high spec way more than you need” game. Android is setup to do that extremely well. So what we do, we provide a highly predictable and sort of balanced proposition where you know that the apps are going to run and you know it is going to be great and there is no risk of a strange user interface someone put on top of your phone. But because of the predictability of our model, we cannot hit the bleeding edge quite so fast. So I think the high-end has also been defined by Android for that set of customers who are fixated on specs.
Do you consider these customers to be a big audience?
It is a big audience, but it isn’t the biggest audience. I do think it is something that gets talked about, so that affects us as well. But really way more people are buying in the range that we are great, way more people aren’t buying the most expensive thing out there. So I think the strategy that we have of being superb from high-middle down to the bottom is a great strategy for growth and that is an enormous number of people are still the minority.
Let’s talk about Nokia being a part of Microsoft. Lots of reports have been written wherein Nokia didn’t share what they were working on with you guys. There was a lot of overlap like Microsoft was working on camera-related stuff and Nokia too was doing something at the same time.
I don’t think there was any problem and I haven’t had the experience of being frustrated that Nokia was ever behaving as a bad partner. They have been a terrific partner and the range of things they have been doing, they talk to us about them all in advance.
So what changes now that they are in-house?
We were in a position before where we had a terrific collaborative relationship where we planned our roadmaps together. They were aware of what we were doing on the OS, and we were aware of their phone lineup. We went back and forth on what would be good investments to make and I expect that to continue. We are trying to do that better and better with other hardware manufacturers as well. In the phone space we had similar discussions with companies like HTC, Samsung and LG among others. So I think that kind of engineering collaboration will remain similar to what has been in the past. It is healthy, it is excellent and it has resulted in great hardware like the Lumia 1020 as well as the Lumia 520.
I actually think the more meaningful effect for us will be in how we go to market. Now we are in a position where we can in a very balanced and thoughtful way can describe the benefits of our ecosystem, which helps all our hardware partners. But also not be colliding in how we describe the benefits of the Lumia phones. So what would happen before was, it was difficult to coordinate where our marketing money would go or what our ads would say. Now I think we are getting much better at that. I think our ability to be effective in retail stores with our messaging and advertising and the way we spend money effectively is probably the thing that will most positively change and help us. I say that because our engineering relationship was already very good.
But at the same time they were also working on Android and just before jumping on the flight I saw an ad for the Nokia XL saying that now you can get Android apps. I have spoken to a lot of guys inside Nokia and the message is people were missing apps, so now we give them an option of Android apps. Doesn’t it undermine all the hard work that has gone from Microsoft as well as Nokia’s side to get the app story right?
I don’t think it undermines it because if you look at the X-series of phones, there are only three parts to that thing. One is, what is the user experience that people are going to encounter and learn as the basic user experience is similar to the Windows Phone. The second is the services it promotes. It is not promoting Google search, or GMS services and is instead promoting Microsoft’s services. The third part is as you mentioned the app platform. It is an unfortunate side effect that those phones don’t also run WP apps, but the fact that they run Android apps is a useful experiment in seeing how consumers react to a value proposition that has two things that are similar and one that is different. You mentioned an ad, and I go back to my first comment about how being one company will help us align with the messaging.
I also remember what you said a day before MWC actually kicked off, and at the press conference you weren’t very excited about them coming out with something like that.
At that time people were asking me a hypothetical question and I gave a vague and hypothetical answer, which people have interpreted in many different ways. The net is that they are doing this experiment that puts out a product with these characteristics and it is interesting to see what happens.
It is interesting you mention that it would have been great if the X series of devices also could have run WP apps. Is that something that can work out?
Unfortunately we are not discussing product roadmaps today.
A lot has been written by Microsoft on their blogs and lot of us too have covered it — the thing with Google and YouTube. How is that going, have you guys reached some sort of an agreement or are they still being adamant?
I can’t talk about what Google is going to do or not do. From our point of view, we think Google has some terrific app experiences to offer to users and we know our customers want them. We are excited to work with Google in whatever capacity they think is the right one and we don’t have any more news on that front.
But is there any headway on that front?
I don’t have any new news on that.
The real WP8.1 devices have only started rolling out, so what is it that you are really excited about?
There are a number of things I’m really excited about. I’m excited to see the effect on the available apps that will happen as the net volume of this new platform ramps up. It will take some time, but once the software developers have the volume opportunities to target the install base of PCs, tablets and phones with one app platform, I’m excited to see the app gap shrink. That is an investment I have a firm belief will really matter, but it is going to take some time.
I’m also excited to see what happens with dual-SIM. You know it is a meaningful gap for a lot of people in countries like India and I’m excited to see how that will affect the perception of Windows Phones. I’m excited about the products that have already been announced like the Lumia 630, the Lumia 930 and in countries like India Micromax has announced that it is coming up with a WP phone, Xolo too is line. So I’m excited to see all these new phones come to the market. I’m also excited to get Cortana going.
Talking about third-party OEMs, is it about filling gaps that you can’t with just Nokia’s portfolio? Do you see them cannibalizing Nokia’s market considering they can bring prices down considerably?
If you say filling gaps, it sounds like a negative thing. I think the smartphone market is so enormous and there are so many different people with different local needs and the range of price points is so great as well. The types of hardware capabilities that end users value or don’t value, it is such a vast thing that even with a wide range of manufacturers all participating in it, in my point of view there is still a ton of room for people to do innovative work and have real businesses that are all meaningful.
You think of this as a giant thing. Nokia has been doing a lot to fill a small part of it, and other people coming to fill in other gaps, but I don’t see them as gaps. There are giant areas of land to go be populated and farmed and mined and that is what I think the situation is all about.
You briefly touched about Cortana. How did you guys go about with the personal assistant feature? Greg Sullivan actually said that you guys saw Siri as Drew Barrymore from 50 First Dates.
We have looked at this area of the digital assistant release over release and we never felt that we could build the level of high quality Windows type experience that we wanted, until now. We looked at what Apple had done. It grabbed people’s attention because it had a personality to communicate with, but frankly it ran out steam very fast. So we had all these conversations with end users who found it to be a fun gimmick, but they never used it. Google has put more functionality in, but it is not interesting.
Because our historic focus has been to be personal and building a smartphone for each of us, we didn’t feel the need to do it until we could put the whole package together. This meant a great voice experience, equivalence between speaking and typing, so sometimes you may be in a meeting and you cannot talk on the phone, so we needed a great keyboard that could do great typing to ask longer things. Great local data, so when you ask questions about a restaurant you would get a useful answer. Knowledge of the web, so Bing has an instant answer for whatever you ask. Bing has had that for quite a while now and the team has continued to invest in it and for a while technically we could not access it via the phone, but now we could. And then there is the personality.
We wanted the full package before we did it and this was the time. So when we started the planning cycle for this release, those were the questions we asked. Could we build the thing that people really want? We concluded that we could and the one constraining factor is to get that value proposition to lots of countries around the world, because people’s spoken accent is different and the local data as well. We think that this is very similar to the Windows Phone story. we have built a vertically great thing and if you live in the US and have a modern phone and try Cortana, you go “Man this is terrific,” but now we have to work to get this thing to lots and lots of people.
Do we see a change where we will have more frequent updates to Windows Phones?
We are on a pattern now where we are doing three updates a year and I’d say in the broad sense that is the pace we are going to stick to. If you define the whole value proposition, there is the OS, but there are also other things like Cortana service in the cloud, there is Bing service that lives in the cloud and there are built-in apps, and we have increasingly moved to a technical architecture and a team structure where we can update a wider range of things on a more frequent basis. So I gave you an answer for the OS, but like in terms of Music, we are updating every two weeks and at some point we will stop updating it every two weeks, but we are currently doing it because we are in the developer preview and there are a number of things that were not as polished as we would have wanted it to be for the actual availability. There are bunch of experiences like that, which we will update more regularly.
What happened with WhatsApp on Windows Phone?
So what happened is that they discovered on our latest update there is a relatively obscure end user setting where users could restrict background data use in conjunction with Data Sense. What happens is if you enable that, you won’t receive WhatsApp messages and you don’t get notified. So they came to us and said that they want their users to have the awareness of that. So we have been working with them on a fix, and I believe they are already testing the fix on beta and I believe they will be deploying it back into the world soon. (WhatsApp is back in the Windows Phone Store.)
You spoke about music, and you have Xbox Music here in the US and in India you have Nokia Music which is massive and pretty well used and is very popular too. So after the integration what happens there?
We are still working on that. We don’t have a plan or an announcement yet, but we are still looking at both the apps to figure out what is the combination of things that can deliver the most meaningful value proposition for our users.
Is there any sort of key trends that you are looking at with the new range of WP8.1 devices that have yet to come out?
I think the one thing that will be interesting to watch will be what happens with our reference design program. There are companies in India and China that are taking advantage of this program to be able to build Windows Phones at low costs and I think it will be interesting to see what markets they are able to be successful in because the nature of that program is where typically the manufacturer is not going to spend a lot of money in marketing. So it might be that they get some traction in the market first that were doing well like the Western Europe, or it might be that they change the dynamics in markets like China. We will have to see how they play out and we are sure to learn lessons as we go along and they will too.
But you ask about trends and hardware and that is an interesting one. In terms of hardware stuff that are coming, I really can’t say.
A big theme has been bringing the prices down, especially we saw Stephen Elop talking about when he introduced the X-series that it is not just this lineup but also the Lumias’ price points that are coming down. Chris Webber also spoke a bit about that. So people who typically buy those products have different needs that can be helped via the OS?
I think those people aspire to the same highly reliable well polished full featured product promise as the people who are spending more money. I think what’s interesting from the OS point of view, how do we tune our software so that we can get those people the most possible thing subject to the missing hardware components. So when you get to that price point, you might not get a forward facing camera, but you still want terrific imaging experience based on whatever camera sensor you have got. So you might only have a 5-megapixel or a 3-megapixel camera or an 8-megapixel camera instead 30 or 40-megapixels, but you still want an incredible easy to use experience with a wide range of apps.
The one other place that as an OS builder we can do that is beneficial to those people is around how data is used. The cost of the phone certainly has an impact on performance, but we can improve that by having a great OS and peripherals like a front-facing camera. But then the other big expense is data, so with features like Data Sense and Wi-Fi Sense, we can help people manage the cost of their data and get the best possible experience out of their phones. So these are the kind of things we think about to enhance our overall value proposition.
The Moto E for instance, which has been quite successful, there was a sense that Android phones at that particular price point don’t offer a good experience. But Motorola sort of changed that. Is that a challenge considering that is a segment where Windows Phones has typically done exceptionally done well?
I think the dynamics are interesting and when we go talk to mobile operators and hardware vendors, one of the dynamics is that people want to see us as a strong third competitor that keeps everyone on their game. You have to ask yourself the question, if the Lumia 520 hadn’t been out with such a terrific low cost experience, would the Moto E gone there or not. I think what we are observing here is that a healthy competitive ecosystem with effective players building great things is good for everybody. And yes, the competition is tough, but we are unperturbed.