Agile is what both the
latest business strategy and software development methodology aspire to be.
The approach promises a significant break with the habits of the past. The ideas are discussed extensively within the IT industry but have yet to become mainstream. Inevitably, it is a case of familiar ideas revamped and repackaged, rather than anything wholly novel.
“There are a lot of companies experimenting with, or about to experiment with, agile software development,” says Steve Gedney, managing director of Borland’s UK operations.
“They want to do it but they’re nervous. It’s so simple in concept that people mistrust it. It all seems a bit touchy-feely…but anybody who has ever practised an agile technology never wants to go back.”
So what is it all about?
Professor Donald Sull of the London Business School has carried out extensive research on the strategic implications of agility, which he defines as a company’s ability consistently to identify and seize opportunities more quickly and effectively than rivals.
That means abandoning the old style of planning for the future. “In many domains of human activity, where action must be taken despite uncertainty about the future, there has been a shift towards a more iterative approach, in contrast to a linear approach,” he says.
The key, he thinks, is the way discussions are structured within an organisation.
Agile organisations have managers adept at leading four critical kinds of discussion: making sense of the outside world to spot opportunities as they emerge; making the right choice from the possibilities that present themselves; “making it happen”; and making mid-course corrections as new information becomes available.
In a soon-to-be-published book* he warns: “All too often, though, conversations at companies bog down in an endless series of unproductive meetings in which the usual suspects cover the same ground without making progress…To avoid that, managers should start by asking themselves a simple question: are we having the right kind of conversation?
“Specifically, are we trying to make sense, prioritise, make things happen or revise assumptions?”
Prof Sull argues that Zara, the clothing chain store owned by the Spanish Inditex Group and headquartered in A Coruna, Galicia, is a powerful example of a company which exhibits agile characteristics.
A “fast fashion” house, in recent years it has overtaken the former market leader, Benetton, by spotting trends and rapidly translating them into low cost, well designed and manufactured garments.
Prof Sull says: “Recently, one of my students showed me a dress made by Gucci selling for €1,200 and one made by Zara for €129. I couldn’t tell them apart.”
One factor, he says, is that Zara owns almost all its stores, whereas Benetton operates a franchise system. The stores act as Zara’s eyes and ears.
“The store managers are talking continually to customers, finding out what they like, why they have brought garments back and relaying this information on a more-or-less weekly basis to the company’s marketing manager in A Coruna.
He also points out that Zara’s design, production and marketing departments are located in the same set of big, open plan offices at headquarters, giving maximum opportunity for conversations and serendipitous meetings.
Garments, furthermore, are produced in Spain rather than lower cost economies allowing new lines to be brought to market with remarkable speed. Benetton’s supply chain is as efficient as Zara’s, Prof Sull says. “It is simply less good at seeing the opportunity.”
Agile software development also rejects the idea of regimented forward planning. It is the antithesis of the linear or “waterfall” approach – which is how much software is produced today and why so much fails to be fit for purpose.
Typically, systems analysts define user needs and prepare specifications for programmers. By the time the software has been coded and tested, the users’ needs have often changed. In the most extreme examples, the software is redundant before it is put into service.
Agile methods are a reaction to the bureaucracy and inflexibility of the waterfall approach. Large projects are chopped into a series of smaller iterations.
Users and programmers meet frequently to discuss progress and make changes where necessary. Change is welcomed. Working software is delivered in weeks, rather than months or years.
Borland UK is an example of a company that has espoused agile project development. Borland today is very different from the microcomputer products vendor of 24 years ago.
Now, it specialises in application life cycle management tools and is well on the way to becoming an agile company in its own right. It was, in a sense, forced down that path through its relationship with British Telecom.
Borland had been helping BT with software development for some years when Ben Verwaayen took over as chief executive in 2002. BT was a troubled company at the time following the hiatus in telecoms markets after 2000.
Mr Verwaayen set ambitious target dates for BT to be market leader in customer service. A new chief information officer, Al-Noor Ramji, was appointed in 2004. He set equally ambitious dates for the IT department to become customer-centric.
Mr Ramji had a background in agile development. Borland, BT’s potential partner in transforming the IT department, had not. Mr Gedney says: “We could not describe ourselves as agile at that time. We were a typical US-owned software house developing programmes using traditional methods.
“When BT started to talk to us about agile, we had to make the decision: we became completely involved or we walked away. BT made clear it was not going to wait six months for us to deliver the next iteration of the software.
“We knew how much investment and energy it would take. We took the decision to go with it and it has been a remarkable journey.”
Mr Gedney says the companies became partners in the project, sharing a high level of trust and co-operation, something rarely achieved between customer and vendor. Mr Ramji was completely behind the development. “We had senior level sponsorship from the outset, which is a critical to success in agile deployment.”
Mr Gedney thinks that 30 per cent of BT’s software projects are now fully agile and that its ambition is to build on that base to migrate agility throughout the business.
“Agile is not an overnight solution. It can be risky without the right planning and processes in place and it is a challenge in a large organisation, especially without a trusted technology partner. However, the benefits far outweigh the time, effort and risk involved.”
Mr Gedney may be preaching to the converted. Finetix, a New York based IT group, specialising in financial services surveyed a number of financial institutions and found more than half had used agile methods and, of those, 73 per cent would use agile again. More than nine in 10 found it a useful or very useful technology.
Variants are appearing. The Dutch company Unit 4 Agresso says its software is well tuned to what it calls post-implementation agility, which is the “capability available to organisations after they have a system installed and running”, says Agresso UK managing director John Crooks.
He points out that flexibility in software is available to everyone but agility is not. Some softwares are so flexible that they are prohibitively expensive to instal or modify after installation, he says, pointing out that most organisations will face change of one kind or another in the next few years.
So is there no downside to agile development?
Michael Homendy, software engineer with the management consultants Robbins-Gioia, warns that “scope creep” can be a real risk. It can be difficult to know when to call a halt to new iterations: “Agile typically costs more than a classic waterfall.
“If an agile iteration turns out to be not what the client wants, you risk losing time, money and large chunks of code.”
He concludes that organisations who simply want to migrate a legacy system to newer technology should stick with traditional development methods. “The risks associated with using agile in a legacy situation outweigh any of the possible benefits,” he says.
* Strategy, Innovation and Change: New Insights on Contemporary Challenges. Galavan, Murray and Markides (eds), Oxford University Press.
Get alerts on UK schools when a new story is published