Archive for June, 2009

Half the money I spend on advertising is wasted; the trouble is I don’t know which half.
John Wanamaker(attributed)
US department store merchant (1838 – 1922)

I’ve been involved in a number of organisations that at one time or another have done ‘value for money’ exercises.  To cut a long story short, I usually have issues with this because, frankly, I don’t think value for money exists:  It is a abstract concept that exists in people minds – rather like ‘the perfect lasagne’ or a ‘brilliant restaurant’.  Paradoxically, that doesn’t mean it doesn’t exist. People have – and should have – very clear and strong ideas about their applications and projects and their relative worth. If they want to describe this as ‘Value for Money’, fine. The problem is that it is difficult to demonstrate it for sure. Like “Tescos’s is better value than Sainsbury’s” (something I’ve heard people have passionate debates about), you’ll never prove it. It’s just instinct im afraid.

We live in a world where people seem to expect absolute certainty and think that everything can be turned into a science: The idea that you might be developing something on the back of instinct and passion rather than a formula the finance people can work out with a calculator will sit uncomfortably with some, but that is the reality – benefits of technology are often intangible and unpredicable.  Get over it.

An example of this came to mind when I was at an event run by a large UK department store many years ago. The web was in it’s infancy and my company (me, really) had done some work developing their first web site. Later, they had added ordering facilities for a selection of their products.

I got talking to a lady that turned out to be very senior in the Marketing area of the organsiation.  Very briefly, she was moaning that although they had invested alot of money into the site (much to my annoyance they had done most of this through a rival company), there didn’t seem to be much evidence that it was generating sales:  The site itself seemed to get alot of hits but the rate of ‘conversions’ (turning these into sales transactions) was quite low.  Overall sales for the company as a whole were doing OK though, but there was pressure to consider the web investment and perhaps even drastically cut it back.  She was dissapointed with what was happening and asked me for my opinion.

“Isn’t it obvious what is happening?” I said.  Silence, shrugged shoulders.  ”People are going to the site, looking up product information and stock details and then going into the shop to buy it” I said.  ”I probably would”. I continued “so the site is doing a valuable job and providing a service (and a return) but not in a way that anyone intended”.

“Interesting” she said.  ”I hadn’t thought of that – how do we prove it?”.  I explained that you can’t and I suppose if I had really wanted to be clever I could I could have quoted John Wannamaker above who famously said “half the money I spend on advertising is wasted, the trouble is don’t know which half”: It’s probably best I didn’t – I always thought it was Lord Lever that said it.

The marketing lady got another glass of red wine and changed the subject.

I have always had a concern about the way User Stories are sometimes regarded as all you need to build something.  Don’t get me wrong, they are a good technique to focuss the mind and provide high-level statements of intent.  The problem is when people do a workshop, come up with a set of stories, and before you know it development is starting.  Very worrying.

Before User Stories become the norm, people used a variety of techniques for expressing these statements (the requirements in other words…) but often it became problematic.  Typically you ended up with granularity problems – some statements represented massive pieces of work taking weeks or months, others were very small, but hard to progress because they had a large number of dependencies.

User Stories at least give an ‘out of the box’ solution to some of these issues, but they still need to be supplemented by something more.

“I think we’ve demonstrated on this project that you can build a system just from User Stories”  One lead developer said to me not so long ago.

“Of course you can” I said. “You’ve been on the project for a year and a half. You’ve had time to build up your knowledge. What if someone new came in with no knowledge of anything?  How confident would they feel, do you think?”  Silence.

He had a point, of course.  If you are woking on something for a long period you have the luxury of getting to know it ‘inside out’. This gives you the ability to ‘travel light’ and also gives you the confidence to think beyond the work itself and to start considering about how you structure and organise it.  The lead developer in this case probably didn’t need the User Stories either, frankly.

Being on a single thing for a long period and getting to know it intimately, is what you would expect in the lean world, given it’s background in production and manufacturing. In production environments people will be working in gangs or on lines doing a single activity. They become experts and then gradually become confident enough to make judgements about how the process itself should be altered.  It occurred to me the other day, that most of the ‘lean’ and ‘agile’ proponents I have met within IT, also tend to be people that are working on single products or single developments and have done so for some time.  They are therefore, to some extent in a ‘comfort zone’.  I never seem to encounter many ‘lean’ proponents that are spread across multiple projects with multiple stakeholders, and that require multiple domain and technical knowledge day to day.  This might be pure coincidence of course, but personally I think it is quite significant.  The people that work in this type of environment are the ones less likely to want to ‘travel light’ and more likely to want a different approach to agile – or rather agile ‘supported by something else sitting outside it’ perhaps.

As I have said previously in this Blog, if you think User Stories are all you need, all I can say is – good luck.