Saturday, June 30, 2007

Star Trek Vocals

image In June 2004, I blogged about a company I'd visited in California, Vocera.  It was early days for them back then but I gather they've now shipped over 100,000 units.  I pitched the idea of using them to a few places - I was very impressed by the product - figuring that UK hospitals would seem them in action and reach for their checkbooks.  It wasn't so and I haven't seen them in the UK - although their website notes that a hospital in Belfast is using them.

Today, loitering by the wine bar at Whole Foods Market, I saw that many of the staff there have a Vocera "badge" dangling round their neck.  I talked to one of the guys about it.  He told me that they loved them - they could call anyone with a single button push and voice recognition, find the nearest, say, "wine expert" that could help them or check the location of a fellow member of staff.   You see this stuff working and you think "wow!"

It's a lot like when I first saw the staff in Wagamama using ipaqs with wireless cards to take orders - the right order straight to the kitchen and delivered back to you a few minutes later. No bits of paper, no scribbles on notebooks, no misunderstandings.

Having spent a little too much time in hospitals recently, I see a big hole waiting to be filled by wireless technology.  I even wonder if police radios could take advantage of the same technology - with private wireless networks put in tube stations so that they could maintain contact and still locate each other, avoiding the need for complicated proprietary technology that is incompatible between forces and doesn't work underground.  If they needed to set up an operations centre, they could just put a wireless node in the middle of where they were and get working.  When WiMax shows up properly, it will be even easier to cover a wider area.

For now, if you want to talk to Vocera about their stuff, you'll need to get in touch with BT or IBM.

Thursday, June 28, 2007

So which do you suffer from ...

iphoria or iphobia?

The web is resounding with talk on the iphone:

Walt Mossberg: "a beautiful and breakthrough handheld computer"

Newsweek: "a significant leap � a superbly engineered, cleverly designed and imaginatively implemented approach"

USA Today: "a prodigy � a slender fashion phone, a slick iPod and an Internet experience unlike any before it"

New York Times: "the most sophisticated, outlook-changing piece of electronics to come along in years"

So that's 4 iphoriacs and no iphobians

But, rely on engadget for a more balanced view.  Some of the things that you can't do with the iphone are really quite stunning:

image - No MMS (so no sending photos by text? good job the camera is pretty average)

- No video recording (huh?)

- No voice dialing (Voice dialing isn't something I've ever got excited about though)

  • - Battery life of 300-400 charges before needing to be sent away to be replaced (at a cost!).  I guess this means "full charges", but I wonder what their life expectation in years based on charging every day or every other day?

- The OS takes 700mb of available memory (make sure you buy the bigger version)

- New wifi networks are announced every time they're encountered (most people will, I imagine, turn off wifi unless they need it, that's how I do it on the phone I have now)

- No ringtones from MP3 (wait for the downloadable ringtones you'll have to pay for)

- No support for Flash (so no youtube?)

- No cut/copy/paste of text (maybe the multi-touch movement for "cut" looked too odd?)

And ... no stereo bluetooth ... so you'll have to stick to the white wire headphones (maybe they should have used a different colour for the phone so that people would know it was an iphone and not just an ipod?)

Still want one? Of course you do. They're going to sell millions and millions purely on the back of those pullout quotes above. I know I still want one. But I think it's safe to short the stock ...

Wednesday, June 27, 2007

British Scheduling

Only we Brits could ensure that the Wimbledon finals, the British Grand Prix, Live Earth, Henley Regatta, the 3rd one day cricket match with the West Indies and the start of the Tour de France all take place on the same weekend, 7th/8th July 2007.

Alternatively, this could be the biggest test of UK logistics ever seen, getting us into practice for the 2012 Olympics.  With 200,000 expected to go to Silverstone, 70,000 to Live Earth, a million or more lining the streets of London for the TdF and then a few thousand more at Henley, Wimbledon and so on, you couldn't ask for a better test.  It's putting an awful lot of eggs in a basket with 5 years to go; but, if you're going to do that, then what better date to pick than the 2nd anniversary of the bombings.  London proves, as always, that it can rebound.

I hope it's a sunny weekend.

Tuesday, June 26, 2007

Things To Think About - Five Years On - For Suppliers

On July 3rd, 2002 I gave a presentation to a team from BT StepChange - a unit that was created to work on e-government (and, presumably, to make a large change, upwards I imagine).   Their web address, found through google at, shows only a 404 error so maybe they're done gone with their changes.

I'd already worked with a lot of suppliers, inside and outside of government, by then and was more than a little jaded.  Every time there was a new project, a bid team would pitch up staffed with some of the brightest people you'd ever met, they'd have references galore for all the great work that they'd done.  You'd poke around and find that the team in front of you had:

(a) never worked together before

(b) hadn't actually worked with the clients that were referenced and, when you took up the references informally, things weren't as great as they'd been painted

(c) wouldn't actually be available for all of the work after the bid, if they won because, of course, they'd have to go on to the next bid.

Maybe I was naive in thinking that such things would and could change.

BT had worked with OeE on the ukonline project which had resulted in UK government's second consolidated web portal (the first was which, whilst loved by many, was despised by just as many - those who loved it were, inevitably, much louder than those who didn't).

If I remember correctly, BT were also running the Small Business Service's or were perhaps still pitching for it during an open procurement.

My jaded self wanted to deliver a few messages about the potential for e-government in the UK and how we shouldn't mess it up ...

BT Stepchange slides - 03.07.2002 - 4 things to do for a supplier 

Some other frustrations were espoused in a slide with these two bullets:

1.The number of technology solutions implemented by departments for identical problems quadruples every 12 months

2.80% of the money on e-government solutions to date was spent on things that the customer never sees

I wanted suppliers to step up, to make step changes perhaps, by bringing new partners to the table who could work together to exploit new technologies and new routes to market - to counter the traditional big supplier tries to do it all and moves like a constipated dinosaur.

I also wanted suppliers to recognise that there was no need to keep building things that had already been built.  Indeed, only today, I read how another department is creating yet another internal staff directory for the intranet.  That will be about the 800th, maybe the 1000th, possibly the 95,000th (globally) that has been created so far.  They're available on the web for $99 and under.  Buy one get one free.

Of course, the jadedness also arose from the fact that all was not well in the world of government IT as it hadn't been for some time - and suppliers were definitely part of the problem (but, of course, not all of it). Some press snippings I collected at the same time (April/May/June 2002):

BT Stepchange slides - 03.07.2002 - slide 1

A full five years on, I'm not sure that the headlines are any better.  There are exceptions, but if you look at the slide above, you'd be hard pressed to say that any of them were completely false - although I'd say that Self Assessment has long since gone out of the headlines, except for reporting once a year how many million tax forms were processes.   That said, I do think much progress has been made:

- Government routinely insists on a "key personnel" schedule in the final contract.  Those who were on the bid team are often held accountable through this contractual device (it doesn't stop them leaving the company of course).  It's a toss up, of course, whether all of those people are the right people to deliver, but it's better to have a team start on day one than have an empty organisation chart to fill.  Suppliers often have to hedge their bets on whether they will win the bid (or any others) and will fill several bids with the same people - their winning more than one bid in the space of a couple of months can really hurt your attempts to establish a solid key personnel schedule

- A full implementation plan is usually insisted on before contract signature is reached, with a very detailed plan for the immediate aftermath of signature (at least giving the delivery team the chance to hit the ground running)

- Detailed risks and issues with the plan are asked for, usually with cost consequences and owners against each so that the government department can figure out (a) how good the supplier is at thinking about alternative scenarios and (b) can see if there's sufficient contingency in the appropriate budget. That doesn't mean that the list is comprehensive or that anyone does anything about them, but before contract signature is a better time to find out what the risks are than afterwards

- Default contracts include templated liabilities, liquidated damages, parent company guarantee and other clauses that help ensure that the risk is borne in the right place.

The Ten Commandments

1 ... Thou shalt not carve thy departmental name in stone

IMAG0002 (WinCE)

Naturally, I don't profess to have any knowledge of "machinery of government" changes that the new Prime Minister might or might not make this week

Sunday, June 24, 2007

Touching Phone

htc-touch This week I've been playing with the HTC Touch phone, retiring (perhaps temporarily my Samsung SGH-i600).  The Touch, as it's name might suggest, is a lot like an iphone, but one that runs Windows Mobile 6.  After a few minutes use it's clear that it is, in fact, nothing like an iphone but that doesn't mean it's not a good phone.  I've lamented how poor recent phone releases have been and how they lack important, nay essential features resulting in much frustration.

First thoughts on holding the Touch relate to its slimness and the good size of the screen, despite the device being very small.  There are 4 buttons in total (answer call, hang up call, power on/off and a volume rocker - that also allows you to switch it to silent/vibrate).  The power button wakes it from sleep and shows you the home screen - the famous one with the weather icons (showing 16C and rainy in London, which would have been pretty accurate for nearly all of May and most of June).  The proper "touch" interface comes alive if you sweep your finger from bottom to top and then from left to right (or vice versa).  It's nice, but not great.  You get access to your most frequent contacts and most frequent functions (mail, text, tasks etc) and then music/photo/videos.

I searched far and wide on the web before I got hold of the phone to see how it handled text/SMS entry.  There are 101 video demos of the touch screen interface but none of day to day text entry.  It turns out that the standard Windows interface is used - you can stab at a tiny onscreen keyboard using a tiny stylus  (that brings back terrible memories of the iPaq), poke at the screen to make scribbles (in palmOS or Sony 800 style) or use that strange Transcriber tool (which actually works pretty well but is slow; and, for it to be effective, you have to write pretty much a whole sentence and then let it transcribe that).  I tried downloading a couple of "soft" keyboards to use instead - hoping to find something that would work with two thumbs, blackberry style.  There are a few that attempt this, but none are much good.  One even lets you rotate the screen to landscape with a full keyboard (unlike the iphone), but even that isn't much help.  Without tangible feedback and the ability to locate your fingers over keys with raised ridges, it just doesn't feel right and any speed increase is quickly cancelled out by increased mistakes.

I also downloaded a threaded text application, Textr, on trial for the next 3 weeks.  It looks to work the same way as the Treo's but seems much, much slower, despite an apparently zippy processor in the Touch.  That might go the way of the dodo in 3 weeks if I can't find a way to make it go quicker.  I'm intrigued why Palm's threaded text application seems to perform so much faster than those written by 3rd parties.  Someone even hacked the one off the Treo 750 and made it available for other Windows phones, but that looks to have been squashed now.

Overall though, there's much to like about the Touch.  Windows Mobile on a full screen with touch input is better than WM on a smaller screen with no touch.  Battery life is good - 3 days standby seems reasonable, with maybe a day and half if you make a couple of hours calls a day.  It's slim and elegant and does all the things you need it to do.  It suffers, in fact, few of the flaws that I've ranted about recently - and it even charges (and uses headphones) via a mini-USB.  It's not, however, a 3g phone so web browsing is average, but there is WiFi if you really need to browse.  The Vodafone Business Email client doesn't work on WM6 so there's no using that yet.

So wither the iphone?  This phone is a good glance at what the iphone will be like when it launches in a few days.  And my instincts say that it's not going to be what everyone wants it to be.  Much as I want one, mostly for the "best ipod ever" functionality (and the sooner Apple release an ipod without the phone, the better), it doesn't look like I'll be rushing to get one, preferring to wait until the end of the year when they come here in native UK version.  I'm hoping that by then, Apple will have implemented a landscape keyboard with wider spaced keys to help improve text entry.  In the meantime, I'm thinking that the best way of paying for one may well be to short the stock on the day the iphone comes out - there's no way that the hype can be justified on release and once "normal" people have them in their hands and see that whilst the ads weren't dishonest, they weren't quite telling the truth about, say, browsing speeds or ease of searching for contacts, we'll see a bit of a backlash that should whack the stock for a short while.  Until Apple updates the software and fixes the things that people don't like and then drops Leopard into the sales channel, sometime in October.

e-Government Do Lists and Lessons Learned

I found out earlier today that Powerpoint is 20.  Microsoft's version of it is a year younger, as they bought the developer in 1988.  Over the 7 years that I've worked in, around and for government, I've done a few hundred presentations, often on e-government, sometimes on project delivery and occasionally on other topics such as cultural change, organisational blockers and such like.  I thought I'd start digging into some of those to see what I said and whether it's still relevant.

From late 2001, I ended most of my slide decks either with a "do list" or with a "lessons learned" - the former was for UK government presentations, the latter often either for supplier presentations or for foreign governments (I reckon we averaged 2 visits a month to our offices during the time of the Office of the e-Envoy).

The first do list I put up, at the "electronic government" conference was on 21st November 2001:

Electronic Government 21.11.2001 - first time list of 9 things to do

Fittingly perhaps, it started with an admonition to narrow down existing do lists - I could already see people building complex sets of inter-related actions to deliver their own visions of e-government.  Experience inside and outside of government taught me that we needed some of the basics first:  XML schema definitions (I'd already worked, very indirectly, on Self Assessment and PAYE whilst at the Inland Revenue who I think were ahead of the game in seizing the XML opportunity), authentication standards (T-scheme was already looking like the wrong horse to back I think), and shared infrastructure (there was no way to implemented joined up online government if everyone had their own systems).

Perhaps the big topic was the buyer/provider point - a re-wording of JY's wholesale/retail model.  Departments needed to understand who they were retailing to and who they were wholesaling to and establish different strategies for each.  I think this point is still outstanding in much of what has been done in global e-government.  Too much is assumed to be essentially retail (i.e. provided at all stages of the process by government) and too little is wholesale (i.e mass market input mechanisms offered to 3rd parties who can wrap their own products and services around them).

I have dozens more of these and they make for cheap blogs, so expect a few more to surface over the coming weeks.  I'm also using a new blog tool that I've found easier than the standard web pages.  It's Microsoft's own Windows Live Writer. I stumbled across it quite by accident yesterday. It's easy to set up, provided you understand your domain name web hierarchy and passwords (I had to look them up of course) and, so far at least, works quickly and effectively.

Saturday, June 23, 2007

Time Waits For No Man

We're all getting older.  No question.  My knee upgrade is testament to the fact that I'm getting older for a start.  At a corporate level, some companies see this "getting older" problem earlier than others.  And whilst there's nothing wrong with getting older, at some point those older people are more likely to leave the company than stay, and that can be dangerous. 

 I worked with a company a while ago where the danger was both clear and present.  I put this slide up at an internal "what are our problems" kind of conference one day:

Ageing Population

In the top left is a graph showing the curve of age (x-axis) versus frequency of that age in the organisation (y-axis).  The average age is marked by the red line.  The blue line shows the average age of the UK population.  This organisation's average age is a little over 7 years higher than the UK's - meaning that they're going to hit the retirement bulge 7 years earlier.  This is pretty typical in engineering-led organisations and is very likely the norm for public sector organisations, particularly policy-led central government ones.

The bottom right is a stacked graph showing the years of service in each age decile.  Pretty obviously, the older people get, the more likely they are to have been in the company a long time.  But what it does show is the years of experience that are going to be lost as the older deciles go into retirement.  In this case:

- 20% of the company's total experience is vested in staff aged 55 and over (and likely, therefore, able to retire there and then under what is often called the "rule of 85", i.e. when years of service plus age totals 85

- 44% of the total experience is held by staff aged from 45-54 with 2/3 of those people having more than 25 years experience

- Less than 5% of the organisation are in their first 5 years of employment

I've seen similar figures for the UK civil service - where 38% of the workforce will retire over the next 6 years and for the US airforce where in the period 2001-2006, 40% of civilian employees were scheduled for retirement. In the US as a whole, 64 million workers are scheduled to retire by the end of the decade - some 40% of the workforce again. Much of this pain, whether the UK or the US, will be felt in engineering industries - oil, gas, electricity, utilities and so on.  Government will be right behind them.

Some of the senior folks in the organisation where I showed this slide saw a looming opportunity - their pension scheme was fully funded and operating costs would fall as people retired.  That's one way of looking at it.  The other way is to wonder what the folks who will soon leave know that the rest of the organisation doesn't know.  In an organisation where things are constructed and operated over decades, not everything happens on an annual or even 5 yearly cycle.

- At times of crisis, it's usually the old hands that know how to resolve the problem.  If you were going in for a major operation would you pick a surgeon who was seeing it for the first time or one who had already carried out two dozen such surgeries?

- Much of the DNA of an organisation is typically stored in the heads of those who have been around a long time.  Few companies that I've looked at really pay attention to procedures manuals, knowledge repositories and so on, unless it's what they do/sell (consultancy companies were amongst the first to make use of tools to capture and distribute knowledge, recognising that it could give them an edge)

But like I said, these folks didn't buy that it was an issue . The idea that they had to worry about capturing the knowledge, putting less experienced employees alongside those about to retire, codifying the information, updating the procedures and so on was just too big a leap.  I called the proposal "Knowledge Continuity". It died on the operating table.

vacheron watch (WinCE)

As a result of my interest in this knowledge space, I'm always on the look out to see who is doing it well, so that perhaps I can learn a better pitch.  So, a few weeks ago I spent a pleasant afternoon with the oldest watch-making company in the world - Vacheron Constantin.  They've been around for over 250 years.  I figured that they'd know a thing or two about knowledge continuity - they promise to repair any watch ever made by them, no matter how long ago.


The first thing I noticed on arrival was that the premises are staggeringly modern - and, indeed, they were constructed just a few years ago.  Their building, on the outskirts of Geneva, is all steel and glass.  Inside, every room is bathed in light.  It needs to be - no point trying to repair a watch part that is no bigger than the tip of a pencil in the dark.

The Swiss watch industry nearly died just a couple of decades ago.  The rise of quartz watches practically killed off the practice of making mechanical watches and it wasn't until Nicolas Hayek (owner of Swatch then but now a multi-billionaire owner of many of the most famous watch brands ever known) started aggregating both movement makers and watch makers that the practice revived.  All the skills were still there, amongst people who had been making watches for decades, in ever smaller quantities, but they had to teach new people the same skills - and that can take several years.

As I toured the "factory" - that word grates on me in this context, perhaps a better word is "atelier" - I saw how they were organised.   The entire operation is divided into discrete stages - a kind of conveyor belt of tasks from polishing pre-made parts to testing the final  fully put-together watch in an enormous machine that appeared to contain room to rotate two dozen watches continuously.

There were "entry level" tasks carried out by the newest (and typically youngest hires) - these folks, both male and female and in their early 20s, polished, shaped and stamped each of the watch components (many are made elsewhere in Switzerland, in the Jura and arrive ready for assembly in Geneva). 

- There were several "training rooms" where new employees were supervised by more experienced workers

- There was a clear progression from entry level to most senior employee. This being Switzerland, I would have imagines much of the progression was age-related, but I was assured that whilst the training programme typically lasted 3 years, there were plenty of opportunities to progress more quickly once that was complete.

The last stage was the construction of one of their most expensive and difficult to make watches - a minute repeater (an entirely mechanical one). I looked at the completed mechanism for this watch under a powerful microscope and was absolutely stunned at the complexity of it.  Each takes 6 months to make and they produce only 2 or 3 a year.  It's made by only one person - so, if you happen to have 500,000 euros to hand, you could say that you knew the man who made your watch.  His name is Christian, by the way, and he's been with Vacheron for 35 years.

Christian showed me all the manuals that they had prepared for how to make the watches - not just the minute repeater, but every watch in each of the collections.  They were finely detailed, showing the secrets of how to make the chime of the repeater just right (it's a secret but it involves soaking it in a special liquid made of oil and some rare flowers).  Every time a new watch was made, it had a detailed manual.  If a watch that hadn't been seen in decades came back for repair, documents were prepared that gave information on how components were reworked.

My point is, these guys knew that their output was likely to be around for years and probably decades.  The Vacheron museum (in the main store in the centre of Geneva) has a full collection including some of the earliest known pieces.  If you're in the business of producing timeless devices (forgive the pun), you want those that follow you to know how they work and how to look after them.  Likewise, if you're in the engineering business - perhaps making clean water treatment plants or jet engines for the 787  - or, indeed, government policies that are likely to be in place for a long time (PAYE is, essentially little changed in 60 years, computerisation not withstanding), you want to know what your people know and how you're going to retain that knowledge.  How you're going to safeguard it so that, when the inevitable crisis comes along, you don't all stand next to each other looking blank.

So you need to know

- What's important to know in your organisation - sometimes that's going to be a house-style of doing something, other times it's going to be a rigid template for a standard design, it might be a problem diagnosis process or it might be just the kind of noise that a minute repeater watch makes when the chime is "not right" (and therefore something that could take many moons to learn)

- Who has that knowledge and where are they on the retirement curve.  If lots of people have the knowledge, it's less important to find ways to record it now.  If only a few people have it, then they're a good place to start testing out your knowledge capture tools - whether that's a fancy system, a paper folder, a drawer full of photos, or a comprehensive manual is neither here nor there to start with

- How do you get the knowledge from that person, using video interview techniques, story-telling workshops, courses where they train the next tier of the team, simple job shadowing, codification in existing systems or whatever

- How do you keep that knowledge up to date and ensure that it's passed on continuously.  How do you spot changes as things develop and need to be enhanced, so that you don't have to go through an intensive capture exercise again

The looming retirement bulge is something that no company has gone through in the past.  This is a one-time effect from the "baby boomer" period.  Our population is ageing, no question, but it's ageing at its fastest rate now.  Any historic turnover experienced by an organisation pales when compared with what will happen between now and 2013.


Speak Your Footprint Machine Launched

Defra launched a beta-version of its carbon calculator this week. There were are few teething troubles on the first day that look to be resolved now. Developed in partnership with the Energy Saving Trust and hosted in a sub-domain of, the site will caculate the carbon footprint of both you and your home. Friends of the Earth, in a giant leap of logic, damned the caculator with faint praise with their statement that although the calculators could "play a useful role" in helping people to target climate change, they should not "distract from the crucial role that government must play in cutting carbon dioxide emissions". Ah, thanks for that, it's always worth being reminded that identifying the problem is not the same as fixing it. The site has some beautiful animations and there's some danger that you'll linger on it, burning ever more pounds of CO2 whilst you study it, but you can whip through in 8 or 9 minutes, less if you have your electricity bill to hand and even less still if you already know how many lightbulbs you have. Someone told me yesterday that they'd entered 92 as their lightbulb count and were expecting a visit from the climate police any minute on the basis that the site asks for your postcode (although to be fair, it only wants the first part - so if the climate police are going to hunt you down, they'll have to check local electricity bills) Intriguingly for government: The software that runs the back-end calculating “engine” is open source, and is, with the support of the Government, being licensed by the software developers, d::gen network limited under “open licenses” such as the GNU General Public License. For more information go to It's only a beta, which means feedback will be gratefully received. Personally I'd like to know more about what my carbon footprint means, e.g. - in the light of coming/not coming/delayed HIPs, what grade would I get? - what is average for my post code? The national average really doesn't mean a lot to me. What about the average for a 3 bedroom apartment like mine? - what is a reasonable target for me to aim for given where I'm starting from - there will be some things that I just can't change or where the business case is marginal (I'm unlikely to rip out the air conditioning in my house - strangely, despite the fact that there are 700 apartments within 500 yards of my home, a/c is not a valid entry for type of heating). Just saying notionally 20% less isn't too helpful And then some additional options: - Add air conditioning, halogen bulbs (my house is full of them and I haven't changed any in 3 years) At the end of the process, I'm told my house emits 9.96 tonnes of carbon a year, my appliances make up 4.6 of that and then there's 5.96 more for travel. That's 15.92 tonnes per year, about 60% more than the national average and 3 tonnes more than my notional target. I erred on the catious side too, so I'd like to think my footprint is at least a little lower. According to the methodology document, 46% of all domestic emissions and 74% of total emissions from home use come from space and water heating - and so the bulk of the calculation must be inferred from my electricity bill (heating is 46% of that I imagine), so I do wonder if it would be different if the calculator knew I had a/c at least (but, there again, it's all driven from the electricity bill - so whether you have a/c, leave your stuff on standby all the time, have 47 fridges or 16 TVs is neither here nor there, given the cost is all reflected in the electricity bill). (Note - I've updated the numbers since this post was originally made, including the total for appliances inside the household emission total) Mr Miliband professes a lower number Earlier, he told GMTV that his "own individual carbon footprint ... was 2.76 tonnes - below the national average of more than four tonnes." I'm guessing that's just his travel - either that or he doesn't eat, sleep or live anywhere. Still, all in all a worthwile exercise. I have an action plan, generic as it is, and I can take an interest in conversations in the lift at work about whether my footprint is bigger than someone else's. If the next iterations add some more features, I'll be back to check them out. And, the site is nicely designed, simple to use and a nice sign a willingness to make use of the branding There are other sites of course, like the "carbon caculator" (which, unlike the defra one requires you to register first). * Disclosure: Whilst I presently work at Defra I have not worked on this project or anywhere close to it.

Wednesday, June 20, 2007

No Fly Zone

This headline caught my eye UAL Corp.'s United Airlines, the world's second-biggest carrier, said a computer failure grounded all its flights worldwide for two hours today United fly around 3,600 flights a day, so 2 hours of downtime would affect at least 300 but probably more given that it would leave flights in the wrong place for return journeys and result in knock on delays with availability of landing and departure slots. It may even have caused delays for other airlines who perhaps had to wait for connecting passengers or who were bumped from their slots to make room for UAL planes. Apparently the system that calculates "weight and balance" failed. I'm intrigued that this appears to be a single global system rather than a rule-based system that runs locally (either locally on the client or locally in a region or country). There was another story that it might, instead, have been the system that records when the planes doors open or shut. That sounds even more bizarre. I thought the cabin crew figured out if the doors were shut. Computer failures aren't just a public sector problem even though the press, certainly in the UK, seem to think that they are. Oddly if you google "public sector computer failure" and "private sector computer failure" the results are roughly the same - about 1,050,000 entries each, but even the "private sector" stories seem to refer to the public sector. That said, I expect neither search string is a particularly helpful one.

Tuesday, June 19, 2007

Re-Delivering on the Promise

I had to laugh today when Tim sent me the press release for Accenture's latest report on e-government. It's called "2007 Leadership in Customer Service: Delivering on the Promise." It's ironic isn't it. After all, my own DotP didn't quite DotP, or at least not as intended. It didn't even make it to its 4th birthday. And yet here we are again with the A folks using the same term (google says there are 324,000 instances of this phrase so plainly there are a lot of promises being made/broken). Apparently they've made it to ten or it might be eight reports, both numbers seem to come up. But it's still unclear what the P is and whether it's being taken or given. The good news is that the UK has leapt into the top 10, 8 places behind Singapore. The report observes, however, that Just under half (49 per cent) of UK public quizzed in the Leadership in Customer Service: Delivering on the Promise report said the standard of government services has remained the same - or become worse - over the last three years. The report suggests governments are being challenged by newer technologies - such as email and SMS - replacing traditional communication channels, and are struggling as they shift their focus from offering front-end services to making sure they are adequately supported by back-end services. The report states: "The time has come to create the infrastructure that closes the loop between expectation and experience." The recommendations for next steps are hardly earth-shattering 1. Build an actionable citizen-centric service vision 2. Build the enabling infrastructure to make the citizen-centric vision operational 3. Build the high-performing workforce that can drive the vision through to fulfillment I like the triple use of build. Not buy. Build. The sub-bullets under (2) are: - Define the processes and workflows needed to reach the vision… - …And don’t wait to get started putting them in place. - Take advantage of service-oriented architectures (SOAs) and shared services as flexible solutions to disparities in government infrastructures It all feels so early 2000s.

Sunday, June 17, 2007

e-Government and piles of fruit

"It's not something we do, it's everything we do" is the tag line that greets you as you enter the new Whole Foods Market store in Kensington High Street. Actually, the whole store is festooned with lines, quotes, lists of values and policy statements, proclaiming the virtues of buying local, the wonders of seasonality, the potential for the Soil Association to do good, the opportunity offered by the employment package (which includes "supplemental health benefits" and stock options). After the signs and the sayings the first thing that you notice about the store is the huge amount of space. This is like an Ikea store, without the arrows telling you where to go - and, amazingly, without the sense of being enclosed in a box. On the ground floor there are no aisles, just piles of food, drinks and assorted non-food items placed almost haphazardly around the place. Of course, I doubt there's anything random about it at all - these guys have been in this business for 30 years and I imagine that they have learned everything there is to know about retailing. The fact that the store was packed on a Sunday afternoon says they're doing something right - and, unlike in, say, Harrods, there were no tourists blithely (blindly?) wandering around gawping at the shelves; these were all paying customers. Waitrose and Marks and Spencers must be quaking - indeed I hear that the Waitrose down the road will shortly close for a major refurbishment (rumour has it they've bought fleets of smart cars to take potential customers to another branch). After you've absorbed how much space there is - you notice this because you haven't been whacked in the ankle by a rogue trolley pushed by someone who you're thankful isn't on the roads - you start to realise how much stock they have and what a huge range there is. A dozen different types of tomato (including some impressive looking black ones), ten different types of eggs (including ostrich eggs the size of a kid's head) and more herbs, spices and other condiments than you can possibly imagine what to do with. With range of course, comes choice, and choice is difficult. Take me to a restaurant where there are 101 main courses and I'll take the fillet steak thank you very much; here I watched people happily stuffing half a dozen different types of everything into bags. It reminded me very much of a market in Marrakech. I suppose this stuff all costs more than it does at Tesco but you'd need to be paying close attention to figure that out - some of the prices look very reasonable (how much is a pint of milk anyway?), others - like the more exotic fruits - probably don't have a Tesco equivalent. And, anyway, it's all guilty conscience money - you're paying a premium, you say to yourself, because they put 5% of their profits back into charity, they help sustain the local economy, they buy their food items from Farmer John in Cornwall who would otherwise struggle to survive and so on. Whole Foods have arrived on the scene with a bang. A 3 floor urban organic superstore, stacked to the ceiling with fresh products chosen from a broad range of suppliers although more often UK sourced than not. It feels like they've taken the "front end" of the supermarket and cleaned it up, creating one of those "retail experiences" that people like to go on about. Everything sparkles, everything is fresh (and date stamped - the oldest coffee had a "roasted on" date of the previous day) and the staff are friendly, seem to be informed and, genuinely, seem to be happy to be at work. Still, it's only been a few weeks. At the same time though, you realise they've done more. They've redesigned the shopping baskets - they have wheels and extendable handles(like those bags travellers from the North West of England to London seem to favour) and are deeper than usual; there are 28 tills with one line for all of them, the tills have a screen facing you that shows everything that has been checked in so far (and at what price), the staff pack all of your food for you (and know enough to put the heavy items in a separate bag from the soft items). And then you realise what they must have done at the "back end" to engineer this set up - how they have built sourcing networks, dealt with transport logistics, recruited and trained a huge salesforce and so on. There were glitches of course, but nothing serious from a customer point that I could see: they have a form for "barcode bloopers" i.e. when an error is made with a barcode or when something doesn't register properly - this must be pretty frequent as I saved £10 with items that didn't have proper codes; there's no chip and pin system yet (seems odd given they've been working on the place for over a year). The oddest thing was that the staff are expected, somehow, to know the difference between any of the near-endless varieties of product - this is no Pret A Manager where there are maybe 20 or 30 things to tell apart, this is a place with several dozen products with 20 or 30 sub-varieties; when there are 5 different yellow tomatoes, all priced differently, you're really going to struggle to figure out which is which unless you're a tomatologist. They'll doubtless fix these issues quickly. So why excited about a new supermarket? Well, this is both "new" and "old" all at the same time. Supermarkets have been around longer than IT systems in government, they've changed little beyond the footprint (Tesco Express and all the other derivatives). This one is de novo for London - setting up a retail chain with this scale capability is no small feat (and whilst I imagine it will be hard to find similar-sized sites elsewhere, I have no doubt they'll start to expand - they have, after all, 190-odd stores in the USA already). But it's also just a "supermarket" - it does everything that Sainsbury or Tesco do today, but it does it all with a twist - it somehow feels "better" than the usual shopping experience. I liked the place so much I even had lunch there. So if something like this can be set up by a new (to the UK) retailer and can have this impact, what would happen if you said "well hold on, government's been in the same business for 20 or 30 years (IT-phase rather than pre-IT), we know everything that can and has gone wrong, we understand lots about call centres and transaction systems ... what if we set it up from scratch like this ... what if we did a First Direct, an Egg, a Whole Foods Market?" Surely you could take an existing operation - a benefits service or a tax credits operations centre - and Whole Food it? I guess it's not that simple. As evidenced by the policies and the values on display, they have thought of things from the top down. Doing this for a government operation would require, in many cases, a rethink of policy - the objectives and the outcomes desired - so as to make them simpler, easier to apply and less prone to errors. At the same time, that simplicity would make supporting IT simpler to develop - IT isn't there to solve complex policy requirements, it's there to make sure the money gets to the people who need it and comes from the people who should be paying it. It is though, even if I say so myself, an idea worth thinking about. Just as I wondered what would happen if Pret A Manger took over customer service in government, it is worth wondering what would happen if a completely new, completely unrelated company took over a government operation. If they had the chance to design from scratch with the sole aim of delivering a range of benefits to those who qualified, I bet they wouldn't come up with 61 different, individual benefits. You know, it could probably be done without any consultants ... all the knowledge already sits inside government.

Saturday, June 09, 2007

E-government is more than just cosmetic

An email forwarded from a colleague today, reminded me of this article that I wrote for Computing magazine in October 2002. The online version seems to have ditched the accompanying photo, for which I'm thankful as I remember it being particularly grim-looking. Government websites need to be consistent, relevant and personalised - not just a representation of the offline world with a thin e-veneer. defines a website as 'a set of interconnected pages ... a collection of information'. By that measure, our e-government efforts appear to be successful. More than 1,800 websites exist in the domain, with perhaps 40 going live every month. But adding 'directly relevant' or 'highly personalised' to the dictionary definition shows that we have a long way to go. Efforts so far have concentrated on replicating what we already have;, perhaps. But electronic government is about transformation. It's not about putting lipstick on a bulldog, covering up the fragmented nature of the offline world with a thin veneer. Consistent and coherent presentation of government information, across all departments, makes for a better customer experience. We must allow customers to easily find what they are entitled to, from where and by when. Today, too much customer brainpower is used to figure out which bit of government does what and how a particular website works. lets you shop in a certain category - just books, say - or search across the whole product range. Whether you are buying electronics or books, shopping with Amazon or one of its sublet stores makes no difference. The experience is consistent, relevant and personalised. Imagine this model applied to government. It shouldn't matter where you are when you look for something. Typing 'child support' into any government site would bring up a set of topics, including details of eligibility, how much it will be and, most importantly, a set of related options. These would be from across all of government, perhaps including child benefit, registering a birth, or applying for a student loan. The tabs across the screen could just as easily be government departments as stages in your life. Achieving this demands content management software. A hundred software salesmen will be rushing to email me with the latest guaranteed one-size-fits-all solution. Sadly, there is nothing 'out of the box' about this. Content management is not something that you buy, it's something that you do and keep doing. It is a business discipline underpinned by strict standards. It will take a big effort to go through all of the information that is already online. Achieving consistent labelling alone is hugely challenging: there are more than two dozen definitions of 'a child', for example. Taking action now would mean a relatively short step to an effective, useful set of websites that present clear information, directly relevant to the customer. I am not claiming that tagging content, loading it into a system and presenting it differently transforms government, but it is a necessary first step. It will allow us to build on all the effort that created the websites in the first place. It is a bigger step, of course, to a 'one click: having a baby' transaction, where everything that you need to register your newborn is taken care of. This is the vision we are working towards and must be in the mind of every government employee working on website design. We may still be putting lipstick on our bulldog, but it will be a much nicer shade.

Reviews, Opinions and Outsoucing Customer Service

I was hunting for a new restaurant to take some friends to the other day. An interesting candidate popped up, but reading the reviews I was left confused. Here's a heavily edit sample. First, the stuff on the plate: "the food was delicious", "Food is rather traditional", "The food was less than average" And the folks that bring the stuff on the plate: "the staff were unnecessarily formal and trying far too hard", "the staff have been trained to be gratingly 'friendly' in the most excrutiatingly artificial way", "the service friendly and unobtrusive" For a second restaurant, I found the following: Food: "Food was good", "food here is, dare I say it, a little more refined", "the highlight was indeed the food", "really tasty, original menu", "The food was first rate" ... sounds very promising, until ... "all taste was eradicated by ginger and lemongrass and the accompanying chipirons were undercooked", "dinner was a warmed up lamb dish and rather tasteless", "standard run of the mill Soho restaurant" and "sadly we only saw a glimpse of the michelin star in the food," And, for the service: "Service is fine bit surely much more than fine is to be expected", "Service was attentive enough without being annoying", "Service was ok, but mostly not very interested in their customers", "service wasn't what you would expect", "Service was generally weak" and then the culmination ... "Service was just rubbish" This is one of the joys of the internet of course. You want an opinion? Take your pick, there are 1001 your disposal. There is, after all, an opinion in all of us - of course, not all of us share ours with everyone (plainly, by writing this blog I don't fit that category). The variation appears not unlike the description of witnesses at a crime scene who tell the investigating officer "he was short", "he was tall", "average height", "heavy set" and "quite skinny" and so on. More than anything, they perhaps say that everyone arrives at a restaurant with an expectation that is drawn, probably, from very few data points and, most likely, no data at all, just a sense of what, say, "a soho" or a "michelin starred" restaurant must be like. All this got me thinking about great "service experiences" and, in the last 9 months or so of working in London again, I think I'd be hard pressed to beat Pret A Manger. Has anyone had a bad experience there? And I don't mean eating a prawn (sorry, crayfish) sandwich despite being allergic to shellfish. Their "locations" (restaurant seems too big a word, cafe sounds an injustice, outlet too retail) are always busy with a constant stream of customers throughout the day, with the inevitable peak between twelve and one thirty or so. There are 150 stores in the UK and they turn over £150 million a year - and I'm guessing they have perhaps 3,000 staff overall. Not bad for a 20 year old business in one of the most competitive sectors in the economy. Every visit is smooth. Staff are friendly and knowledgeable - despite what must be an enormous turnover given the vast variety of nationalities represented at the counter (does anyone else try and guess which country someone comes from by looking at their name badges, proudly displayed on their baseball hats? No? Thought not). I imagine Pret to be a place that those looking for a few weeks or a few months of work go to as they pay for a trip around the world, pay their way through college or, perhaps, as they start out in their working life and want to see what's out there - the training period is 10 days (probably more than most first-timers get?). You also have to "fit" - one in seven applicants gets a job. At the same time, for those who want to stay with the company, they're committed to developing you and offering wider opportunities. Mystery shoppers keep everyone on their toes and, if service is good, staff are rewarded with bonuses (Pret say that 80% of staff get that bonus every week). Pret is a company with no marketing or PR department and one that doesn't run advertisements, except for posters in its own windows. Sathnam Sangera, from the FT, worked there for a day - which is perhaps a little PR-like. So what would happen, do you think, if we gave the Pret management team the job of running customer service operations in the public or private sector? If we took those contracts away from Big Company A who know just as much about service as the next Big Company B (which is perhaps to say not as much as they should) - and here I define BCA and BCB as the stodgy, large bank balanced, able to pay liquidated damages-types - and gave it a small, innovative, leading-edge company that believes in itself? Sure it would be a near impossible scaling up of operation at the highest end but, my point is, if the average call centre operation has far fewer than 500 people (and, in many cases far, far fewer), what would happen if you injected the Pret delivery style into a different evnrionment? Could an organisation from such a different background teach the big public sector behemoths how to run their service business? Running call centres is a very tough business - I've watched it up close for some of the last couple of years, sitting in on calls, listening to irate customers (occasionally even listening to customers delighted with the service offered). If I worked at Pret A Manager I'd also spend at least one full week a year working at the front line - which is not something I imagine we could say for most managers in customer-facing organisations. The prankster in me thinks of the opportunity to rebrand government services - Pret A Payer (Tax), Pret A Donner (tax credits, pensions), Pret a Changer (transformation teams) and so on. But I can also that more than a few cynics might go for Pas Pret A Donner - an do forgive the destruction of a beautiful language there. There's no easy way to create a great customer experience but it seems to me that Pret has succeeded in a business where expectations and opinions would normally lead you to count against it. Isn't that just like the public sector problem? P.S. If anyone has read this far and is wondering if I went to either of the restaurants above, I went to the second, Arbutus, where I had a stunningly good meal in a boisterous, loud & buzzy restaurant with wooden floors, small tables, bad art on the wall and, probably the best innovation I have seen in a restaurant in recent years, a wine list that lets you select any wine to be served in a small carafe - whether it's £100 a bottle or £20 a bottle. Service is what I expect in a place like that - it's busy with high table turnover and the staff are running around a lot sometimes taking their time to get back to you, but I really don't mind pouring my wine myself (in fact, I probably prefer it - it lets me set the pace) and I don't mind gaps between courses (I'm almost never in a race to get out of a restaurant - I rarely go to eat before I go somewhere else). So, visit the place and see if you have a different experience.