The first principle of the Agile manifesto states: ‘Our highest priority is to satisfy the customer-through early and continuous delivery-of valuable software.’ Incremental and iterative delivery is fundamental to all agile methods. That is, rather than waiting until the whole, complete product is ready to launch, delivering small parts of the whole product early and often.
Delivering early allows us to check our assumptions and test out innovation - are people really prepared to rent out a room in their homes and will others pay to stay there for a night? It radically alters the business case by bringing revenue in sooner and improving cash flow, plus being early to market can help with both share and awareness.
Most elements of software can be delivered separately and still have value. A complex software application has value before it is complete, automating a single part of a process can still be helpful, for example. Software is also inherently malleable. It can expand and contract, be reconfigured or changed or updated.
But people find it hard to think this way. Products and ideas often feel indivisible. ‘I need it all’, ‘it’s just not valuable without <insert feature here>’, ‘our competitors have that feature, we need it too’, ‘it’s more efficient to do it now’... the reasons for keeping things big are endless - and often they are hard to resist because they seem persuasive.
The only real counter to this is to present with equal firmness the benefits of early delivery and the dangers of the big bang.
Case Study: Boo.com - The lure of the big bang
Boo.com is famous. Or perhaps infamous is a better term. During its glory days, this sports/fashion online retailer, was named one of ‘Twelve Start Up Superstars’ and was lauded in The Wall Street Journal, the Financial Times, Fortune magazine and Vogue.
Now the company is on other lists: ‘Boo and the hundred other dumbest moments in e-Business History’ or ‘The top 10 dotcom flops’. The founder’s own book is called Boohoo: from concept to catastrophe.
For anyone who was asleep or at school during the dotcom boom and crash, boo.com was seen as embodying the excessive consumption and management errors of an era.
They burned through $135 million of investment in 18 months, (to be fair, Ernst Malmsten, the CEO, denied charges that they wasted money on champagne, by pointing out he only drank vodka). There were usability troubles with the website and glitches with the technology platform - the fancy graphics and 3D imaging was unbearably slow at a time when most people had dial-up internet connections.
Flaws in the business model, like the number of returns, undercut expectations. Boo.com had a vast staffing cost, employing 400 people in multiple countries around the world. Because it was launching in numerous countries with their own tax laws, languages, and fulfilment methods, the system set-up complexities were enormous. Dealing with these delayed launch and required vast technical support.
What was the real problem?
A blog from former employee and IT Director, Tristan Louis, commented that the biggest lesson was: ‘Plan early, think of all that can go wrong, and then plan it again. Usually, spending more time on specs saves you from many headaches down the road.’
It’s a tempting conclusion - but it’s wrong. The biggest error was not delivering in increments. As Ernst Malmsten admitted, ‘instead of focusing single-mindedly on just getting the website up and running, I had tried to implement an immensely complex and ambitious vision in its entirety.
‘Our online magazine, the rollout of overseas offices, the development of new product lines to sell on our site - these were all things that could have waited until the site was in operation.’
These things could certainly have waited, but the website that Malmsten envisaged was still fronting an overambitious model. Incremental delivery would have highlighted problems earlier, from the user experience to returns.
By the time boo.com collapsed, it was turning over $500,000 in orders every two weeks with good conversion rates and a healthy repeat purchase rate. These were great figures - but too late. A revenue stream that had come in earlier, even a smaller stream, would have off-set the huge quantities of investment being poured in.
The appeal of big
So why hadn’t Malmsten considered incremental delivery? Did he just not think of it? Possibly. But there were several reasons why Malmsten and his advisors were seduced by the big bang.
Suppliers - who boo.com needed to win over in order to have something to sell - were attracted by the idea of a global company. They weren’t interested in dealing with an internet company that simply seemed like an alternative distribution channel. Investors also wanted to deal with a global company.
The team were worried about competition. They wanted to raise the barriers to entry so that no-one else could copy their ideas, (they should have been worrying about how easily they could be asset-stripped when they went bust).
These factors led the team to reject geographical incremental delivery, launching country by country, and instead to launch in many at once. This led to enormously increased complexity in the back-end of the business in order to sort out taxes, delivery and pricing.
Boo.com envisaged a near-instant global dominance that would discourage competition, create a ready-made image of cool and could be extended swiftly to other market sectors. In giving way to this ‘big win’ temptation and rejecting incremental delivery, they lost everything. It’s a reminder that business as a whole, as well as IT projects or software products, need to stay focused on delivering small, valuable increments early.
