Wednesday 31 March 2010

Number 35 in Wired UK Britain Top 100 Digital Power Brokers

The messages had been coming in for several days now - and today there was an avalanche of them so I knew it must have arrived in the shops. So I succumbed and bought a copy.

"It" is the May 2010 edition of Wired UK - and "I" appear to be Number 35 in Wired's first annual survey of Britain's Top 100 Digital Power Brokers.

I should say straight away that "I" should be "us". Tesco and Tesco.com is packed with people doing innovation. I've been in R&D for some years now and so how come suddenly outsiders are taking notice now?

The answer is simple: Tesco had been doing innovation on the quiet creating and implementing amazing new ideas but not saying anything about them.

Tesco is a brand that is 'admired but not loved' - yet behind our corporate machine image is a whole raft of people totally energised by a desire to try really hard for our customers. Why? Because we enjoy listening to customers and what they like/dislike about the Tesco experience and strive to improve the service as a result - not forgetting that we are customers too.
  • We build what customers would like to use to have an enjoyable experience at Tesco - for example, find products easily whether in-store or online.
  • We improve technology in store because we go and work as staff in stores at least twice a year and get annoyed by tasks that technology could improve or simplify.
  • We instill the best security practices because we don't want our own personal details seen or our own payment details compromised.
So one day I decided we should shout about this stuff and not keep quiet; have the boldness / madness to wear innovation on our sleeves in public - and be totally, breathtakingly honest, complete with documentation of failures and more.

So this blog was born, and with it our successes and failures on the path to bring customers great new products and services.

It's worked, too! Birth of our API was a resounding news story, our TJAM Innovation Day was covered by the media, and Apple rejecting the first edition of our Tesco Finder app made it even to non-IT magazine The Grocer. Only this week, Computing magazine’s front page spoke about the launch of Paypal’s API by referring to ours (complete with a picture of a Tesco.com delivery van). So outsiders now do think of us as innovative. Yes!

One final point: I can do a lot of things quickly in R&D to 'prove the point' and even take projects to trial with customers and staff. But the reason I can do this - and a reason 'I' managed to get to number 35 in the Wired UK chart is because of colleagues around our business who build reliable, scalable production systems. Colleagues who have the grace to let me stand on their shoulders.

Sunday 28 March 2010

Tesco Internet Phone service closes at the end of April 2010

Tesco Internet Phone, a service set up by Tesco Telecoms in 2006, is to cease operations on 27th April 2010.

In a statement issued on their website, Tesco Telecoms say:
From 27th April 2010, we will be closing our Internet Phone and Talk Wi-fi business. This is due to a number of factors that impact the service we provide to you, so with regret, we have decided to discontinue this offering.

From the 27th April 2010 you will no longer be able to access your on-line account or be able to make calls using your Tesco Internet Phone. If you are a Talk Wi-fi user then you will no longer be able to make calls using the Talk Wi-fi application on your mobile phone.
Tesco Telecoms, like Tesco.com, has learned the art of being outstandingly innovative for customers, and wanted to try this 'unknown' area to see if they could find a way of making it a commercial success by providing a simple, easy to use internet telephony service.

Tesco Internet Phone was launched at a time when making phone calls was rather more expensive than it is today. At the time, internet telephony was a somewhat 'geeky' experience mired in complexity and not friendly to non computer-savvy users. Tesco's entrance into this area was considered to add a sense of credibility to the industry as well as being simple to install and use.

However, Tesco Internet Phone's arrival also woke up many telephony providers to review their offerings and pricing, and the result has been that the main reason for customer's choosing our service has ebbed.

In addition, many internet telephony providers now offer a whole host of value added services. Look at Skype with its many innovations such as Skype-To-Go where you can apply for one or more free dial-in numbers in cities around the world and use Skype to forward your local call to an international destination at rock bottom prices. Skype continues to lead the world of consumer-based internet telephony through investment and innovation and has my total respect.

Tesco Telecoms' decision to close their service is partly an acknowledgment that internet telephony has moved on from their offering, and that they can concentrate their resources on their other products such as their award-winning Tesco Mobile service.

If you are affected by the pending service closure, please read through details about this service by accessing the Help section of the Tesco Internet Phone site.

Friday 26 March 2010

Mobile Phone App Stores - Handset Manufacturers vs Cellular Operators

The following post is my own opinion, not necessarily the opinion of my colleagues or anyone else at Tesco. I should also point out that I am project lead for R&D (Research and Development) at Tesco.com, not all of Tesco as some of the media have stated!

---

Now maybe it's just me. Maybe I don't "get it". Maybe someone could enlighten me?.. but..

What is the point of a mobile phone 'app store' run by a cellular operator?

As we build up momentum with our mobile strategy, you can imagine that our primary focus is ensuring that our web site will work on the majority of mobile phones. I know that's stating the obvious and I hope to goodness that if you're a web site owner you have a similar strategy in place.

On the other hand, there are a few web site services at Tesco.com that deserve the 'special experience' that a fully-formed application can bring - Grocery and Tesco Clubcard being prime examples. The decision to write a fully-formed app rather than adapt the web is based around how important it is to create a best-of-breed experience on mobile phones that takes account of the challenges of being mobile. For example, the grocery application will be useful because it will allow customers to build up a basket without needing a constant internet connection (say, on a train journey with those annoying signal-free tunnels).

If we write an app rather than adapt the web, then we need to code for particular makes and models of phone, because each handset can have a different operating system, screen size and shape, and amount of processing power. Yes there are 'app adapters' where you write an application once and it can be compiled to work on several handsets, but if we are going to make the 'app' effort then we may as well do it properly and write well-targetted code.

So, we need to work with the manufacturer of each targeted handset model to perform the development process, from obtaining the software development kit right through to placing the finished app in their App store.

The nice thing about this process is that we also get engagement from the handset manufacturer and can access technical support (and I dear-say some marketing opportunities too). The real 'win' is that we can then reach any customer using that particular handset, on whatever cellular network they reside.

All handset manufacturers have - or are getting - this development / support / marketing infrastructure in place, visibly seen by customers through App stores - Apple App Store, Nokia OVI, Android Market, and others.

All this is great: If we want to target, say, a Nokia N97 because we like the interface and think customers would get an engaging experience with a Tesco application on it then that's a decision we can make, and work with Nokia to ensure that all customers who have an N97 can get the app and enjoy a great experience. Placing the app in the Nokia OVI store seems obvious to us - it will reach all Nokia customers with N97s across the UK on whatever cellular network they happen to be on. "Lesser" phones won't see the app on OVI because the system detects their phone and filters out apps that won't work.

So please, help me understand why we would even think about putting an app into a cellular provider's app store?
  1. Why would we only target a certain cellular provider's users? We want to reach all customers! Do we then have to go round to all stores? Yes - but would be easier targeting the one store created by the handset manufacturer.
  2. What about new unlocked handsets (not tied or branded to a cellular provider)? They only have a choice of their handset manufacturer's store anyway.
  3. What if customers change cellular provider? Do apps downloaded from one store get barred from working? Even if they do keep working, is there a possibility that a customer may think they will stop, so feel they are locked in to their current provider (and maybe resent us for apparently forcing this situation?).
The fact is, you need a handset for mobile communication so why not work with the people who built that handset? App stores provided by hardware manufacturers offer a simple message of 'these apps will work on your phone - and we know it because we built your phone and we've tested the app!'

I think cellular providers' app stores will confuse the message for customers and I'm prepared to say so publicly because I'd like to see a debate started somewhere so I can understand whether cellular app stores are any more than a desire by providers to somehow get in on the game for the sake of it.

In my opinion, cellular providers don't want to see the elephant in their room which is that they are becoming just a way of connecting to the internet. They're trying to think up ways of being more than an internet service provider but actually that is what they are. The 'power' is in the handset app/web browser and the content service providers offering web sites and services.

It's not the first time they've tried this - in 2005 I was invited to speak at a conference about mobile payments (where you pay by mobile phone and it goes on your cellular bill). I warned the organisers that I'd be a black sheep and even forwarded the slides of what I was going to say, but they accepted me anyway.

I told the conference that Tesco has a perfectly fine payment system in place on the website already, nice and simple for customers to understand and that all that was needed was to have an internet connection from handset to our web site and customers can type in their credit card just as they always have. Oh, and that every single shopping website in the world works like this - no need for any cellular financial intervention, thank you.

I'll never forget spotting the furious shaking of the head from the back of the room from one of the sponsors of the event. Everyone else was nodding.

I said it then and I'll say it now: All customers need on their handset is an internet connection between their app and our service - and nothing more.

I'll add: Given apps need to target a particular handset model, who is best able to get that app to all customers in the lest confusing manner - a cellular provider or the actual manufacturer of that handset?

---

If you are a cellular provider with an App Store, I would be happy for you to guest-write a post for this blog with your counter argument which I will publish un-edited (unless it's an ad of course!). Contact me at nick@lansley.com to take up this opportunity.

Thursday 25 March 2010

Azure to the rescue

Developers following the build of the new Tesco Grocery API, (an R&D project that allows third-party access to our grocery service through their own applications) may well have questioned a consequence of enormous download counts for our iPhone Clubcard app.

The question: If we had similar downloads for the forthcoming Tesco iPhone Grocery app, how will the API cope if it is still in R&D?

The blunt answer is that it won't. My tests have shown that currently the API service can survive processing around 400 simultaneous HTTP requests on its server - any more and performance begins to suffer - not very useful if several thousand customers all start trying to synchronise their favourite products (a key feature of the grocery app) at the same moment in time.

What to do? Productionise the API, of course! Well that work has already started but it's going to take a while to complete, and we want to launch sooner rather than later.

The good news is the API has been created as an ASP.Net application - perfect for being hosted within the new production Windows Azure cloud computing platform. An early version was running in an equally early version of Azure during 2009, so the design takes account of operation within the cloud. It's also been upgraded recently to work in .Net Framework v4.0 and ongoing developments use the Visual Studio 2010 dev environment, so it's 95% cloud-ready right now.

Importantly, the API does not itself store any personal information since it acts as a proxy between client applications and the new 'Project Martini' grocery service running on the live Tesco.com web servers. That's important as it would be difficult to describe where the Azure SQL databases actually reside to the Data Protection Registrar, so thank goodness it doesn't need any such registration.

Tomorrow it's a case of adjusting the API source code to work fully within Azure's development fabric, then upload it to the Azure cloud for some stiff weekend performance testing. Indeed I need to turn up the stress on the API until it starts suffering in terms of performance, then I'll be able to calculate how many CPUs and other Azure resources are needed to scale the service upwards to cope with possibly thousands of simultaneous users.

Cloud computing is a great way of 'taking the strain' with services - particularly with peaks that will no doubt be experienced after an app launch - I'll let you know how the Azure-hosted Grocery API behaves over the coming days - and post launch.





Wednesday 24 March 2010

Apple iPhone adverts double Tesco Finder download rate

Tesco Finder, our store and product finder app for iPhone, has been featuring in Apple advertising on the back of the Sunday papers over the last four weeks (thank you Apple!).

You may be interested to know three interesting statistics:
  1. Tesco Finder downloads on these Sundays have been double the normal download rates (and remember, each download is a new customer using the Finder service). In recent weeks Tesco Finder has secured an average of 760 downloads a day, but on the days the advertising has been shown, this has reached over 1,300 - pretty much double.
  2. Use of the application (defined by the count of calls made by Tesco Finder to our servers to perform the search requests) increased by 20% on advertised days, returning to normal the following day.
  3. On advertised days, our other apps also enjoyed an uplift: Tesco Wine Finder downloads increased from 145 downloads a day to 245 on advertised days (approx 60% increase), and Tesco Clubcard app increased from an average of 10,000 a day to just over 13,000 on advertised days (30% uplift).
Reviewing these stats, I draw the conclusion that Apple's featuring of apps as part of their advertising is highly effective at both bringing new users to an application, and also reminding existing users to start using an app they may have stopped using recently (or perhaps use it more) judging by the uplift in use on those days. It's also interesting that both downloads and usage returned to normal the next day once the awareness spike had passed.

There is also a 'halo effect' uplifting other apps in our collection in the app store. What makes this all the more interesting is that, before the advertising went live, we agreed with the Apple agency to change Tesco Finder's icon to show the full Tesco logo rather than the previous image of a "T" with a compass. Although the uplift in Tesco Finder was most effective, this may have helped our other apps too: People just glancing quickly at the advert would note the familiar Tesco logo and note that "there is a Tesco app in the app store" and search on just our name.

I would be interested to know if many (perhaps all?) iPhone app downloads increase on advertised days. I would also be interested to know if paid apps have the same percentage uplift as free apps do.

Sunday 21 March 2010

Google Nexus One - an R&Dist's delight (Exciting but Buggy)

I’m just back from my visit to the USA and, as promised, I used the Google Nexus One as my main mobile phone for most of the visit. This post reveals my experience (in comparison to my iPhone 3GS), and it is one of mixed news I can tell you!

In summary, the Nexus One is very much a geek’s phone, full of exciting features - but does not have the polish of an iPhone experience. It is also buggy - to the point where I had to restart it once a day unless I used it just for voice calls and text messaging - hardly the point of a smart phone.

In order to fully immerse myself in the experience, I pre-ordered a USA Pay-As-You-Go (PAYG) SIM card l so that I could play without incurring any nightmarish roaming charges. Such SIM cards are freely available on eBay and I arranged with the seller for it to be delivered to my first hotel, which happened without any incident. The seller even activated the card for me so I could discover my USA mobile number and send it to those who needed to keep in touch before I arrived.

The Nexus One quickly connected to the network and my voice calls and texts behaved as normal. Email and browsing worked fine too, and the phone locked into the hotel’s wifi network easily - informing me that I had a ‘login’ page to go to before it could continue and starting the browser to help me.

The phone’s News and Weather app (which consists of a widget on the phone’s home screen with the weather symbol, temperature, and a randomly selected news story, plus a full app with all news stories) quickly realised that I had left London and switched to the local weather in Bellevue, WA. It also adjusted its “UK” news menu to “US” to give me local stories.

The Google Map was also surprisingly accurate, even if I was deeply inside a building. There would be no hope of GPS there so the phone must have triangulated its position based on the various AT&T cellular towers as well as remembering its last GPS fix (perhaps assuming I could not have strayed far) because often the central arrow on the map was exactly right even if the phone was showing a large “margin of error” circle.

I was able to receive emails and send them out easily enough, but the more I used the phone the more I noticed that the ability to work out where my fingers were on the screen became inaccurate. Over a period of a couple of hours of continuous use the degree of error became worse until completely different parts of the screen were responding to my touch, and the keyboard became impossible to use as the error was at least two keys “above” the ones I wanted using the on-screen keyboard. The only way to resolve this was to turn the phone off and on again. I found this happening frequently throughout my trip: the more I touched the screen, the more the calibration went “out”. Has anyone else found this?

The Nexus One’s apps, as found in the Market app, are a geek’s joy - and I think this is where the iPhone and Android audiences diverge. I downloaded the following apps - all free:
  • Barcode Scanner - useful for, well, scanning barcodes using the phone’s camera - quick and accurate.
  • ConnectBot - great for Telnet connections using a command prompt.
  • Facebook - great app for keeping up to date on Facebook BUT when sending images they end up a lot smaller resolution on Facebook than the iPhone equivalent.
  • Flip’n’Shake Lite - an app that uses the orientation of the phone to perform various tasks. For example, if I’m on a call and wish to switch to speaker, I simply move the phone from my ear and place it upside down on a table (so the speaker is facing up).
  • Fring - a way of using Skype albeit with lower voice quality. There is a Skype app but it doesn't work at all well on Android (come on Skype!).
  • Google Goggles - scan anything and it works out what to search for - hugely impressive and a pointer for my desire to “scan the product and understand it, not its barcode”.
  • Google Sky Map - superb app for identifying stars and planets in the night sky where you are - a total joy to use and really inspiring when presented with the clear Nevada night sky during my trip.
  • Google Translate - say a few words and Google Translate converts them to a different language and says those words in that language - closest yet to Douglas Adam’s “Babel Fish”.
  • Layar - great augmented reality app - I have on my list the task to create a Layar data-set with all the world’s Tesco stores and their locations on it!
  • Listen - a podcast app, excellent for picking up my favourite BBC podcasts such as “Friday Night Comedy”.
  • Network Discovery - find out more about the other hosts and devices on your connected network - a geek’s joy.
  • Newsrob - a Blog reading app designed to work with Google Reader (my preferred blog access site) by downloading articles for offline reading.
  • Remote RDP - Windows Remote Desktop - works very well despite the small screen!
  • TasKiller - an app that kills other tasks - VITAL! (more on this shortly).
  • Twidroid - a simple Twitter client.
  • Wifi Analyzer - analyses Wifi signals in order to lock onto a desired signal, and for suggesting the ideal channel for a home Wifi access point by seeing what other signals are on the various channels available.
My PAYG SIM card offered me 100MB of data for the week, which I thought was more than enough given that I would be within wifi range for much of the time. Wrong! The reason:

Background tasks! Whenever I ran Facebook, Twidroid and other apps and then exited them… I hadn’t exited them! They became background tasks (in default settings) so they could keep my incoming tweets and friends’ Facebook updates up-to-date. The apps didn’t care whether I was on wifi or cellular - as long as they could see their respective servers across the Internet they went about their business silently - eating into my cellular allowance in the process.

This is the downside of apps as background tasks - on the iPhone no 3rd-party apps are allowed to remain in the background - to kill the Facebook app is to truly end it. Moving from my iPhone mindset to the Nexus One cost me in terms of the amount of cellular bandwidth. Once I used up my 100MB allowance, the background apps crashed through the rest of the money I had put on the SIM card account at the rate of $1/MB and soon I was incommunicado!

Overcoming my $10 dollar loss, I researched the problem and installed TasKiller which reveals all the background apps, and I killed the ones I didn’t wish to keep running. Indeed there’s a satisfactory “kill all” action which ends all but the most vital phone background tasks (such as monitoring for incoming calls). It even wipes out the animated background screen. My second week in the USA was cellularly inexpensive as a result.

Finally, the Nexus One lacks the finesse of the Apple phone. It works well but I need the geek in me to help understand what's going on. Just my opinion but it’s little things such as the icons not as “polished”, through to there being less inherent sense of reliability. This phone is an R&D’ists phone compared to the iPhone’s production quality. I’m in R&D and instantly forgive its quirks. Others will not be so patient.

Some of you may say that it was unfair to go from iPhone to Nexus One - it’s bound to compare unfavourably because it is so much more advanced. Maybe, but you have to be able to pick up a phone and use it - not fiddle about with it or restart it every couple of days. I’m sure that some bug-fix / feature-enhancing updates are on the way (after all, that has happened with iPhone several times) and I look forward to it.

Having said that, the Nexus One has a great character and form, and will be a significant platform for smartphone technology both now and in the future.

I’m glad I own one to see what happens.

Monday 15 March 2010

A future TescoPlex? Headline thoughts from a visit to Microsoft and Google's HQ campuses.


Forgive me; currently I’m suffering from brain-fade.

I’ve just completed an intensive week at the headquarters of Microsoft (Redmond, WA) and Google (Mountain View, CA) learning how they “do” innovation as well as meet with the right people who are helping us build and confirm (or divert!) our future strategies such as in 'mobile'.

Learning how both companies set about creating and implementing new ideas was an exercise in imagination and impulse; not being afraid to fail, and giving individuals the autonomy - and accountability - for innovation.

Microsoft provided a mature, reflective view on innovation. Meeting with people such as Scott Guthrie (with whom I presented on stage at PDC 2008), and other leaders both at Microsoft and Google provide such insight into the need for - and thought processes behind - innovation.

One thing that completely struck me was the fact that both Redmond and Mountain View campuses felt like university. Lunchtime at Spitfire, a restaurant at the heart of the Redmond campus felt like an upmarket student eatery - and beyond the building the sound of a lunchtime band playing and beer being enjoyed evoked the recall of my days at the student’s union.

At the GooglePlex, as Google's Mountain View HQ is known, an intense sense of energy was palpable and the 'university' campus was even more real with students... sorry I mean employees... accessing all kinds of social functions available whenever they needed it. The mindset of Google’s two founders, Sergey Brin and Larry Page, is everywhere. For example, no desk is further than 100ft from a canteen or food/drink station. Food of the finest quality cooked by top chefs is available in the central canteen - and everything is free. An onsite gym and “endless” swimming pool are available (where you swim in a straight line but remain stationary to the pool edge as you control the speed of the water flow past you). You can get a haircut, a massage, and even counselling for every aspect of your life. Sit on a loo in the staff toilets and the seat is heated. As you sit, you can read door-mounted A4 sheets of paper on good testing practices (“Debugging Sucks; Testing Rocks!”). No toilet rolls but you press a button which controls a water- and air-directed cleansing process while you remain seated… now that was an experience!

From my observations, the psychology behind both Microsoft and Google campuses is that employees make it their home. Every need is taken care from breakfast to evening meal and entertainment - and everything is free. You just go home to sleep, and then you come back for breakfast.



Tesco at the Googleplex: (left to right) Laura Wade-Gery, CEO, Tesco.com & Tesco Direct; myself; Ian crook, Marketing Director, Tesco.com; Angela Maurer, Senior Marketing Manager, Tesco.com; JJ Van Oosten, CIO Tesco.com / Tesco Direct / Tesco Mobile; Julia Tishenko, Category Marketing Manager, Tesco Direct; Kendra Banks, Category Marketing Director, Tesco.com; Graham Harris, Category Director for Tesco Entertainment and Electricals (missing from the picture is Barney Burgess, COO of Tesco.com Grocery & Wine and Rob Salter, Category Director for Entertainment).


At reception is a screen showing the latest live Google search requests.

If last week wasn’t enough inspiration, this week I am at Microsoft’s Mix10 in Las Vegas and I’ve just seen a great keynote speech from Scott Guthrie on Silverlight 4 (write once, run on Windows, Mac, Windows 7 Mobile and more). Mix is for another blog entry later.

Tuesday 2 March 2010

Apple use Tesco Finder in their iPhone advertising campaign



I've just received the artwork for the latest Apple iPhone campaign, and so I can reveal that recently Apple contacted us to ask if they could feature our R&D iPhone app Tesco Finder in their latest poster / newspaper advertisement (see above - closeup of Tesco Finder entry below).

I said, "Yes".

Actually, being ever the consummate professional, I may have toned my response down for your benefit. In reality it was more, "YES PLEASE OMG THANK YOU!!!", knowing me.

Thanks Apple :)