Agile & Lean in a World in Recession Date:   Sunday , June 07, 2009 The world economy has been captured by a recession and probably, as many claim, the biggest one since World War II. This has had, and will continue to have for a period of time to come, a significant impact on organizations in their ability to do business as well as the way business is done. Companies are facing declining revenues and increasing and fixed costs. This has resulted so far in thousands of people becoming unemployed. Ways for significant cost reduction need to be established. But not only the workforce needs to be slimmed down; operational processes need to be tailored and optimized as well to support this reduced capacity within the current economical situation. Traditional, waterfall based processes do not have a track record of being cost accurate (leave alone being cost effective), nor do they have a good reputation for delivering high value for money. Despite the fact that waterfall thinking is ingrained everywhere in our thought processes, the above pain points have resulted in more and more organizations searching for and adopting new styles to improve cost accuracy and their net business value. This is one of the reasons why ‘lean’ and ‘agile’ have seen rapid growth since the previous recession at the start of this century, and why there is so much interest in these concepts in the current recession (12K hits in a Google search using the words ‘agile’ and ‘recession’ in the past one month). Making it all ‘Lean’ Again Lean software on one hand recognizes the fact that many software packages have been extending functionality over time to remain competitive and attractive, and also that many of these additional features often had no immediate compelling incremental value to the intended end users. This has resulted over time in feature rich but often overly complex solutions, with a big percentage of functions never used in practice (excluding during vendor demonstrations). Complexity and expertise go hand in hand, so with complex solutions you get (expensive) technology experts. You see these, for example, with most of the ERP or BPM vendors today, positioned to ‘complement’ their offering and ‘maximize’ time-to-market. With the ‘Lean’ trend some vendors are providing simplified versions of similar offerings, but concentrating on ease-of-use (so that, for example, technical business people can do the work, saving a lot of money avoiding continuous expert consultancy) and maximized value. A good example might be the new wave around Lean BPM solutions. These solutions are designed to automate and audit business processes while keeping design, implementation, control, and maintenance simple in comparison to traditional BPM offerings. Maximize Value with Minimized Effort using Agile On the other hand, you have the ‘agile’ wave taking an alternative route to produce results. As explained by the agile manifesto now already around eight years ago (http://agilemanifesto.org/), 17 pioneers of agile found alternative and better ways of developing software and bundling their thoughts and experiences they came to value: * Individuals and interactions over processes and tools * Working software over comprehensive documentation * Customer collaboration over contract negotiation * Responding to change over following a plan In essence, what these mean is to go back to the core of software development again (or become ‘Lean’ again): * Constant customer or stakeholder involvement in software projects to ensure that what’s being built meets the demand * Increasing the value a team can deliver by removing waste (things a developer usually does but have no immediate value to the customer or stakeholder, but are adding extra cost to the project). * Removing the silos by creating cross functional and empowered teams with high levels of communication and collaboration to ensure everyone is on the same page and is accountable for the results. * Producing working versions of the software that are demonstrated and discussed about with the end users to ensure regular value monitoring and allow for inclusion of constantly changing (business) requirements. This is especially important in a time of recession where today’s environment might be very different from that of tomorrow. All the above points resonate very well in a climate of high recession. No wonder this reaction against the waterfall approach received so much acceptance and attention towards the end of the previous recession in the beginning of the 20thcentury. Let’s take a closer look at the benefits, when costs need to be kept low and value needs to be maximized with the limited (often reduced) resources: * Agile promotes working into small teams (7 plus/minus 2) and the team being collectively responsible for the work they take on board. Some agile practices promote pairing, where two developers work together on one machine when they write code. This comes in handy in case workforce reduction is mandated. The risk of lost intellectual property and transition time is kept to the minimum. * Agile promotes short iterations (2 to 4 weeks typically), often called ‘sprints’. How much and what kind of work will be part of each iteration will be a group decision at the beginning of each cycle. If we take SCRUM as an example, all things-to-be-done will be added to the products backlog. Each item can be estimated in a variety of ways and, as all processes should be kept ‘lean’, only in ways that add value to the goal of producing quality results to the business stakeholders. Often estimates on priority or value and effort or complexity are part of the mix. At the beginning of a sprint the team can decide on the amount of work (read backlog items) it will commit to, often taking it from the list in business priority (value) order. So, where is the value here in times of recession? It is simple; as iterations are small in duration and as the backlog typically is reduced in order of value, when time needs to be reduced to save cost an organization can see results at the end of each sprint (another agile practice) and make the decision to stop or continue on another sprint. * Agile recommends active end user involvement in the sprint teams. This approach of close end user involvement makes agile (together with the short iteration cycle mentioned above) truly ‘agile’. Especially in a market that will change rapidly as a result of various side effects of a recession (team reductions, company acquisitions, and market demand changes, to name a few) only the end users will be in a position to know these business changes rapidly and will be able to discuss them within the development project at the right time. In traditional waterfall based development project, once the set of requirements are determined (often a lengthy process on its own), any changes made to them a) require a change management process to be followed, and b) will cost the project more effort and money with the increase of the impact of change on a work already done, not to mention the plans that have to be revised because of the effect of the change on the existing plans. This last one should not be underestimated: often hours of planning time go up in smoke when substantial changes cause development plans to change dramatically. Using agile practices, any change to already completed work will become part of the backlog and is estimated and selected the same way as any other item. A change to something still in the backlog is costing no extra effort other than adjusting the item in the backlog. Change is king in a recession, and within agile change is embraced, so again a natural fit. * Agile promotes people and interaction over processes and tools, and as mentioned before teams are kept small in size. Often these teams work together in the same room, making interaction extremely easy. In times of recession travel expenses often need to be kept to the minimum, so working from home to avoid commutes gains in popularity. So, what could be the effect on teams having to operate and collaborate when they are no longer located within the same space? Again, simple, while not promoted per se, if tools are needed to support the process, they will be used. Most of the tools in support here are exploiting the Internet, as people work from their homes and can only use the Internet to collaborate, or are even Internet based, or ‘on–demand’ as it is often called (saving on capital expense, often important in times of recession). When ‘agile on–demand’ was typed without the brackets in Google last month it produced close to 47K entries on the subject. Summary and Final Comments It is clear from the above that in times of recession ‘agile’ and ‘lean’ will continue to generate interest and gain in popularity. Does it mean that we will see a big boost in agile adoption now? Probably not. The reason has more to do with the change in approach and people involved in this change process. It is a fact that organizations are risk averse when they need to be extremely careful about cost. Given that most organizations worldwide employ resources that have been grown up with waterfall thinking, it is not easy to convert them all into agile teams. Will the organizations be willing to take the risk to change their practices? Will the individuals be open to change their practices into something they are not familiar with: this ‘agile’ thing? Will the individual be capable of adopting the new agile principles? The last question is not so simple. Agile puts a lot more responsibility on the individual and the teams, so there is no more hiding behind the project manager’s bad plan, the end user’s wrong requirements, and so on. This level of commitment is something that will attract some resources, while scaring away many others. The potential of agile and lean to help in times of recession is there for all to see, but much will depend on how much an organization and its resources are willing to commit themselves to make it a success. Eddy Pauwels is Director, Domain Experts International, Serena Software