Public Strategy

Sometimes self-service is hard

Once you have got your customers into a self-service channel, it's quite a good idea to try to keep them there.  Sometimes things go wrong, sometimes systems are down, sometimes users do unexpected things.

So you need error messages.  Ideally ones which tell the user what is wrong, what they need to do to fix it or when the system will be responsive again.

Error : TICKETBOOK_EXPIRED somehow doesn't quite do the trick. 

The block capitals and the underscore signal pretty clearly that this is something which has floated up from the depths of the system and was never really intended to see the light of day.  But here it is, untamed and untranslated.

Perhaps they want more calls to their helpdesk.

The intelligence of the crowd

Twenty years ago, I had a problem to solve.  A museum was creating a database of the objects it held and needed to be able to produce reports in catalogue number order.  The difficulty was that different numbering systems had been used at different times and catalogues with different numbering systems had been amalgamated without changing the separate numbering systems.  The database needed to include objects acquired and given a number over a hundred years ago, and objects acquired yesterday.  Some numbers had the year of acquisition as a prefix, some had it as a suffix, and some didn't have it at all.  Some numbers were just numbers, some numbers also had some letters.  Some numbers had hyphens in them, some didn't; hyphens appeared at different points in different kinds of numbers, and meant different things depending on where they appeared, and so it went on.

For the curators, none of this was a problem.  Given an arbitrary list of catalogue numbers, they could always tell you what the ‘right’ order was.  They were completely consistent about it and their approach was perfectly logical - but it had very little to do with the ASCII ordering the database program liked best.

So my problem was to write a routine which would  parse a catalogue number entered in free text form, work out which of the various possible sequences it belonged to, format it to look right to the curators, use that version whenever the catalogue number was displayed, create a separate version of the number across several fields which could be used to generate an index which then came out in the order the curators expected, all without requiring them to enter the numbers in the first place other than in the way they had always written them.

It took me a day or two of thinking and a few more days of coding and tweaking, but I got there.  I was quite impressed with myself:  I had no training or experience of this kind of thing, and had had to work it all out from first principles.  The curators weren't impressed in the slightest:  the database just behaved as they expected it to.

Twenty years later I have a different problem.  I am confronting the wisdom of many crowds distilled into the departmental expenses system.  It has long been the mantra of Oracle and SAP that because their installed base is so vast, their software represents the best way of doing things, distilled from the experience of millions of user interactions.  The organisations which use their software should therefore adapt their processes to fit the software rather than have the temerity even to consider adapting the software to fit their processes, as the software embodies the best possible way of doing things.  Government departments don't take that kind of challenge lying down, of course, so it doesn't always turn out that way in practice, but the principle is clear enough.

So let's start by entering the time into a time field.  Simple: 

0600

Or, perhaps, not so simple:

Please enter the time value 0600 in the correct format HH24:MI

Let's try putting in the date:

13-05 is not a valid date.  Please re-enter.

OK, let's try it a different way:

130508 is not a valid date.  Please re-enter.

There are things humans are good at, and other things which computers are good at.  Putting simple data elements into standard formats is, without a shadow of doubt, something for computers, not humans.  What's particularly irritating about these error messages is that they are half way to solving themselves.  It would take so little to go from saying that 0600 should be in the format hh:mm to simply putting it in that format.

So the interesting question is why there is no attempt to be more helpful.  I don't know the answer to that, so I can only guess.  My guess, though, is that the underlying reason is that their users are not their customers.  I can't walk away from this system.  It wasn't chosen because it meets my needs.

That's all a salutary reminder that the same is true for the systems we impose on our customers, which makes it doubly important that we take the trouble to establish what they need and how the experience will work for them.

But enough of this distraction, back to expenses.  Or not:

Unexpected URL parameters have been detected and will be ignored.

Which I think means, you haven't been using this for a while, so you have been logged out and will have to log in again.  But that isn't quite what it has managed to say.

Lean, efficient - and obsessed with customers

Peter Day's latest programme for the World Service is about the principles of lean manufacturing applied to service industries.  The download is here, but probably won't work for more than a few days.  The link to the programme web page is here.  As usual, he finds some interesting people to talk to.

One of them is James Womack, one of the authors of The Machine that Changed the World, an early account of lean in Japanese car manufacturing, who talks about why so many call centres are horrible:

Well the view is that to take people seriously costs too much

If you ask any company that has people waiting on hold or waiting in any kind of queue, “Why do you have people waiting in a queue?”  and the answer is, “Well it would cost too much to deal with them instantly.”

Queuing is always costing money to the provider organisation.  It's costing a lot of money that they don't really quite know how to account for. 

A bit later, there is a manager of a call centre run on lean principles:

I can tell good customer service from bad customer service immediately now.

As soon as I get told I am wrong about something, I know that they are not applying lean.  And if I get a very quick answer, I know that they're not applying lean.

And after a quick foray into the lean-based customer obsessiveness of Amazon comes a man from GE Money:

All your marketing leaders and marketing managers will probably have satisfaction surveys that they used many times and over many, many years, and we have proven our satisfaction surveys really are not linked to change, growth or volumes...  

We are using something called net promoter score, which actually is asking your customers, “would you recommend us to a colleague or friend?” and asking only two or three questions, not a twenty five line survey that a call centre somewhere asks.  This is about getting direct feedback from customers and actually doing something about it. 

It's linked very much to trust.  If you are recommending a product you have, it's much more of a significant link to growth opportunities for a business than “yes, I am satisfied with a product” ... a personal recommendation is much stronger, and if you link that recommendation to your lean philosophy, what you are saying is, “I am listening to you and then we are acting and doing something about it”.

Womack rounds things off with a great provocation:

Just ask yourself what would happen if the provider had to pay the customers' value of time.  Do you really think you would be waiting in all those queues? 

This is all salutary stuff because it really brings out the fundamental link between customer service and lean, which is too often obscured by lean being thought of as being purely and directly about efficiency. 

I had not heard of net promoter score before, but it turns out to be a well established idea with a web site of its own.  The explanation is very simple:

There is one question that provides the best predictor of customer loyalty and for the vast majority of business: How likely is it that you would recommend (Company X) to a friend or colleague? The "Net Promoter Score" or NPS, is simply the percentage of customers whose answers identify them as promoters minus the percentage whose response indicates they are detractors.

The concept of recommendation clearly cannot translate directly into a public sector context, or anywhere else where there is a monopolistic provider, but the distinction between a relationship and a level of trust about which people are willing to speak positively to others and more passive satisfaction is a critically important one for any service provider - we may just need to think about quite what question we want to ask.

Of course whatever the question, if there is no means by which the answers are heard and responded to, there isn't much point.  But that's another challenge.

Use and usability

After an afternoon spent confronting an application from which the third panel could be a screen shot, this rings horribly true.

Even better, in a deeply perverse kind of way, than the wry smile the cartoon prompts, are the 135 comments it has provoked on the post where it first appeared.  The amount of passion unleashed by a simple cartoon is quite remarkable, including some very defensive reactions which are extremely illuminating.  They fall in three main (but not entirely separate) groups:

• It's OK for Google and Apple, they can afford to spend vast sums on making it pretty, I had a small budget which just covered basic functionality
• It's an unfair comparison - my company needs very structured information, and you can't get that from a single data entry field, whereas Google and Apple are both single simple functions
• There's nothing wrong with complexity in an internal application:  it may be worthwhile investing in learning the complexity if that's going to make you work more efficiently for several hours a day.

I have a smidgeon of respect for the last of those arguments.  There is something impressive about the unix command line die hards who need just six keystrokes to send their laptops in orbit round the moon.  More generally, systems which allow power users to accelerate through an application rather than slowing them down to some lowest common denominator version are good, not bad.  But that's not at all the same as saying that users should be required to be expert to counteract the underlying unusability of an application.

There are two ripostes, also to be gleaned from the comments:

• Just because an interface looks simple, it doesn't mean it's not powerful
• The fact that you can make your internal users find their way through an inefficient and unintuitive system, doesn't make it a good idea to do so.

And buried deep in the comments, someone quotes the great line from Antoine de Saint-Exupery:

Perfection is achieved, not when there is nothing more to add, but when there is nothing left to take away.

(via Creating Customer Experiences, via Dominic Campbell)

Innovator's irony

The Innovator's Dilemma is a book, published just over ten years ago, the central argument of which is that it is very hard for dominant firms in a market to innovate radically, because by doing so they risk destroying their existing business.  If they don't, though, they make themselves vulnerable to some upstart which doesn't need to worry about the past and which invents a new way of doing things -and so the original company gets destroyed anyway.  Which is what makes it a dilemma.

The first example explored in the book is the development of computer hard disks - from the earliest refrigerator-sized 5Mb disks onwards.  To oversimplify horribly, at each stage of significant and disruptive technical innovation, new entrant firms dominated the new technology, leaving the established firms of the previous generation to keep squeezing performance out of the earlier technology.

Seagate was one of the beneficiaries of this process, riding the wave of the 5.25 inch disk in the 1980s, and thereby ensuring that none of us has ever heard of the companies which dominated the 8 inch disk industry of the 1970s.  Seagate was in turn slow to move over to the next generation of 3.5 inch disks, but was quick enough to survive and eventually prosper again.

But now along comes the most disruptive challenge to the hard disk industry of all.  For the last fifty years, hard disks have been increasingly super-charged gramophone records:  at their heart, there is still a real disk rotating very fast on a real spindle.  That's not the only way to store data, as the memory stick revolution shows, but until now, solid state drives (which have no moving parts) have been too small and expensive to replace traditional hard disks as the main storage device for a computer.  Now that's changing, with real advantages for users as a result.

Seagate's response is to threaten to sue all the new entrants for patent infringement, while insisting that their existing market is not threatened.

None of that has much obviously to do with public strategy.  Except that if a major and innovative player in an industry characterised by rapid change and by innovation cycles disruptive enough to change the landscape of the industry on a fairly regular basis, and where somebody has pretty much written the textbook on what the problem is and how to manage through it, still trips over exactly the same problem and still finds it hard to respond to the market signals anticipating change, just how surprising is it that governments fumble with the same problem?

How far away is the new world?

Martin Stewart Weeks has a question. 

For the most part, government is not being done in recognizably different ways and certainly not ways that, in any reasonable interpretation of the word, would count as ‘transformation’ . Underlying structures and systems remain largely unchanged. It’s even harder to discern much real shift in underlying culture and behaviour. The compelling and provocative insights of the kind of world that Charles Leadbeater sketches in his new book We Think (as just one example of many analysts and writers) - open, collaborative, subject to new forms of collective intelligence, breaking down the silos etc - are notable too often for their absence in the settled and intractable world of public policy and public management. 

Am I being unfair?

I think the short answer is ‘yes’.  It is unfair, because there is an unspoken switch of focus in the comparison.  No, there isn't yet much shift in underlying culture and behaviour in public policy and management.  But that is surely because the public sector largely reflects the wider society of which it is part, not because it fails to.  Most organisations do not live in a web 2.0 world.  People reading blogs are a small minority, people writing them are a small minority of that minority, and the conversations we have among ourselves are not representative of how the rest of the world thinks about and does things.  For the most part, business is not being done in recognisably different ways - or at least, it is instructive to think about where it is and isn't. 

Take the level of resistance and incomprehension shown by the film and music industries.  The signals for change have been clearer, the potential rewards for change much greater, alternative models far more obvious and key parts of the market much more ready to change than is true for most public service providers.  Napster showed the big labels what the future looked like in 1999:  they chose to look resolutely in the other direction.  Mark Cuban dared to challenge the idea that DVDs should come out months after cinema releases and later suggested that the best thing to do with film soundtracks was to give them away - and is treated as an idiot by the rest of the industry. 

Those are interesting examples for a couple of reasons.  The first is that both industries are inherently digital, so might be thought to be more likely than others to recognise and respond to the challenge of the changing environment.  The second is that in both cases they have embraced changes in the means and methods of production while resisting changes to the means of distribution:  the dynamics of the back end and the front end have been very different. 

Government is, of course, about more than production and distribution.  It is about engagement and participation too.  Most of us are free riders there, just as we are with wikipedia and linux.  Amazon and Netflix will use their customers' actions and opinions to improve the recommendations they make - and in a delightfully recursive way, Netflix is using the wisdom of the crowd to improve the way it uses the wisdom of the crowd - but they don't want us to talk back.  I am struggling to think of a mainstream commercial service where the principles of marketing as conversation are really at the heart of the approach. 

That leaves social media.  It's pretty self-evident that government is not much like Facebook.  But in many ways it would be odd if it were:  government is not about democracy, it is the thing to which we apply democracy.  Facebook is more a substitute for an evening at the pub than it is likely to be for an evening at Conway Hall.  It is a good thing that No10 is playing with twitter, but right now that's on the margins of the margin, and it is probably more important that Patient Opinion is doing what it does - which of course is not a government service at all.

Governments are as vulnerable to the innovator's dilemma as anyone, so it is not altogether surprising that this should be so.  And it is made more difficult again by the fact that it is rarely if ever straightforward to innovate in government services without the consent and participation of government - which is essentially the point Tom Steinberg is reported as making in Martin's follow up post - so reducing the power of the disruptive newcomers to replace conservative incumbents by innovating round them.

So is government slow to change and to embrace the new opportunities created by its changing environment?  No argument there.  Is there greater shielding from the effects of that environmental change in government than in other sectors?  Yes, probably (though that needs to be a description, not an excuse).  But is all that a sign that government is operating in some different and more antique world than other large, long-established organisations?  That's much less clear.  As I have reflected on before, there are reasons for change in government being slow which aren't intrinsic to its being government.

Making government more invisible

People in government often like to think that government is important, failing to realise that they are seeing it from a rather unusual point of view.  For most people, most of the time, government is most successful when it is least visible, or perhaps least intrusive.  It's nice that I can report the holes along the pavement in my road, which somebody dug three weeks ago and then abandoned, through FixMyStreet, but I would really much rather not have to.  Filling in my tax return online is much easier than wrestling with the old paper forms, but I would just as soon not have to do it at all.  I still marvel about the presentation I heard, probably five years ago now, lamenting that it was difficult to get people to become familiar with government online services because they didn't need to deal with government all that often - and suggesting that the solution was to find ways of getting people to contact government more frequently.

Paul Johnston reports on the rather different approach being taken in Liverpool, listening to a presentation by David McElhinney, the chief executive of Liverpool Direct:

David’s focus was using customer contact to drive up performance and drive down cost and at the heart of his philosophy was the paradoxical assumption that customer contact was essentially a failure, something that should be systematically managed down in many areas to an ideal target of zero. On this approach generating a large amount of customer contact is a reflection of excessively complicated processes imperfectly implemented. So the solution is to simplify the processes and move towards one single way of doing any generic activity - so one way of receiving payments whatever those payments might be, one way of making payments, one way of collecting debt etc.

Of course, it can't always be quite as simple as that, but focusing on the right number of contacts provides a hugely powerful way of identifying where problems are.  In the examples I started with, the right number of contacts for reporting holes dug on behalf of the council to the council is self-evidently zero.  But the right number of tax returns going from me to HMRC is one (at least from their point of view).  You can't claim a benefit without claiming it, just as you can't vote without voting.  Even some of that is about timescales, of course - we could have a tax system which didn't need tax returns, but given the tax system we have got, some people have to do tax returns - there is an analogy here with the idea from economics that fixed costs are only fixed for certain periods or ranges of activities, and that in the long run all costs are variable.

So the more interesting form of the challenge is to identify and eliminate redundant contacts, perhaps by starting with a strong assumption that all contact is redundant and then asking what needs to be added back in order to add to, rather than subtract from, total value.

One obvious way of reducing redundant contacts is to reduce overlap and duplication.  There is often an assumption that that means integrating everything with everything else.  I have often wondered just how many people in a normal year want to renew their car tax, apply for free school meals and obtain a licence for burial at sea as a single transaction.  Understanding which services (from the providers' point of view) cluster together to address which needs (from customers' points of view) is pretty important. 

Liverpool seem to share this approach:

[David McElhinney] argued that the council does not really have an interest in having a full picture of its relationship with individual citizens, since if a customer gets in contact to complain about their rubbish bin not having been emptied, they are unlikely at the same time to want to discuss the recent planning application they made or their application for a resident’s parking permit.

The clustering Paul describes in Liverpool - CRM for payments in, separate CRM for benefits and payments out - sounds a bit odd in principle (the distinction is quite a producer-focused one), but probably works well enough in practice.  Worth finding out more.

Let's not delude ourselves that channel shifting is easy

Even when the new channel is literally a couple of steps away:

I went to the movies yesterday...

So, the place is packed. Lines for the ticket counter is 25+ people deep.  There are 6 ticket vending machines sitting there, happily waiting dispense tickets. Empty. No lines. No people. No usage.

I waited and watched for about 15 minutes. 3 people used the machines.  Nobody came to the slow moving line and suggested they use a machine. People were in line complaining about how slow the line was moving.  People were using credit/debit cards in the line so no excuse there.

As you think you will change people's habits with your stuff; go the movies.

I went to the movies yesterday...

So, the place is packed. Lines for the ticket counter is 25+ people deep.  There are 6 ticket vending machines sitting there, happily waiting dispense tickets. Empty. No lines. No people. No usage.

I waited and watched for about 15 minutes. 3 people used the machines.  Nobody came to the slow moving line and suggested they use a machine. People were in line complaining about how slow the line was moving.  People were using credit/debit cards in the line so no excuse there.

As you think you will change people's habits with your stuff; go the movies.

Drop the front page

I hadn't realised just how soon after yesterday's post a perfect example would come my way.  An email this morning from Government Computing, albeit from an email address any ruritanian spammer would be proud of.  Not just a circular - this is personalised (small hint though guys - using the format ‘Dear PUBLIC’ doesn't look quite as personal as some other ways you could have done it).  Let's not quibble though - they are eager to tell me that Guardian Public has published ‘a special independent report' called A Journey Through Service Transformation.  It sounds enticing - and instant gratification is to hand:  ‘to download your free copy... simply click here’ it says.

So I did.

That takes me through a redirection page, no doubt to log my click against the mailing campaign, to a page on the Oracle site, which tells me about the ‘FREE exclusive report’ all over again, with a big red graphic imploring me to ‘DOWNLOAD NOW’, and in case I somehow miss that, a text link telling me to ‘Download your free copy here today’.

I am already thinking that if they were really that keen for me to read this report they could have let me have it by now, but the chase is on, so I click on the next link in confident expectation of finally reaching the prize.

Curiously, this gets me to yet another site which, a quick whois search tells me, is run by an outfit called Marketing Options Ltd who seem to live in Weybridge, but that probably doesn't matter very much because the page is still resolutely Oracle branded - ‘Download your exclusive FREE report today!’, they enjoin, in bigger, redder text than ever before.

Well I thought I was.

But no.  There is nothing whatsoever on this page which lets me have the report they have been dangling in front of me.  What there is here is a form to fill in.  ‘Register below to download your FREE exclusive report’, it says, in considerably smaller and greyer type.  And if I were to fill it in, I could click the button at the bottom marked ‘Register’ and maybe, just maybe, like a rat navigating a maze, there would finally be the reward in the form  of a report or perhaps a small piece of cheese.

I don't know, though, because I didn't go any further.  Nobody at Oracle, or even Marketing Options Ltd, thought it worthwhile to tell me what attraction to me there might be in registering or what might happen as a result, and I get quite enough multi-channel junk mail already without adding to it.

So what has all this achieved?

• I have wasted some of my time.  That matters to me, but not, apparently, to Government Computing, Kable, Guardian Public, Oracle or Marketing Options Ltd.
• My opinion of Government Computing and Kable has been diminished:  I don't recall inviting them to send me third party spam.
• My opinion of Guardian Public has been diminished:  a report only available on an Oracle website, accessible only after registering with Oracle does not come within my understanding of ‘published by Guardian Public magazine’.  They may have been paid to write it, but that's not the same thing.  That may be grossly unfair:  the report may be brilliant and genuinely independent in its content, but I am never going to know that, so that just leaves them with a bit of reputational damage.
• My opinion of Oracle has been diminished:  they seem to be playing bait and switch, and have succeeded in taking a positive opportunity - they could have made it possible actually to get the report and to have got a pale glow of the reflected glory from the ever so independent Guardian Public - and turning it into self-inflicted damage, because they couldn't resist the temptation to play games.

What a result for one small email.  The Robert Niles story I linked to yesterday was about a newspaper so wrapped in advertising that it was impossible to find the content:

Whenever I open our door on Sunday mornings, I never see the Los Angeles Times flag staring up at me from the porch. Instead, I see a two-pound advertising circular that, I know only from experience, contains the LA Times buried somewhere within.

Good service and good design both start with respecting the user.  This is a text book example of how not to do it.

Hold the front page

From Robert Niles via Doc Searls:

Readers owe you nothing. They have no responsibility as citizens to read your reporting, and no responsibility as consumers to look at your ads. They have the right, and ability, to go about their lives without ever once glancing at your publication.

If you want people to read your publication, you then need to do whatever is necessary to make them want to read it.

That means leading with your best shot.

He is talking about newspapers - and more particularly US newspapers - but the general point remains.  Design a website, or a service more generally, for the benefit of anybody other than the user of that service and if they have a choice, they won't.  In many government services - and not only those - that gets easily obscured because in practice users often don't have a choice, at least in the short term.  In the longer term, they could in principle find themselves a more user-friendly government, but the feedback mechanisms there tend to be weak and indirect.  The more likely outcome is that the credibility of government as service provider takes another hit and the apparent attractiveness of other ways of getting things done goes up.  That may or may not be a good thing, but it needs to be clearly understood as  part of the price of poor service.