Eric Ries is Right that 5 Why’s is Difficult
On my continued journey through Eric Ries’ book “The Lean Startup”, I am now at chapter 11, ‘Adapt’. In this chapter, Ries posits that not only should be product be adaptable, but so should the organization as a whole. To be adaptable, the organization needs to be a learning organization, and the way to do that is to learn how to solve problems. Ries points to a well-established Lean problem solving technique known as 5 Why’s. Actually this isn’t so much a Lean problem solving technique than something all 5 year-olds are born knowing how to do – ask “why” until either the root cause is found, or dad says “because I said so”.
There are of course a few other tools that are handy to have in one’s arsenal, such as Ishikawa Diagrams (fishbone analysis, which force analysis on problems based on a set of predetermined categories of causes – for manufacturing these are ‘Equipment, Process, People, Materials, Environment, and Management’). There is no denying, however, that your 5 year old kid has a point, and 5 why’s is, when done well, incredibly effective.
Ries spend some time explaining how difficult 5 Why’s is to do well, and how a facilitator familiar with the method is an important part of success. Why is this so? Why is something that a 5 year old can do, so tough to apply meaningfully in business? Ries talks about the danger of the process becoming a blame session (the 5 Blames) when individuals, and not processes and systems are focused on. He also talks about making sure the problem under investigation is as defined as possible, is not too big, and is currently relevant. Other reasons that Ries does not touch upon are the issue of “Symptoms in Disguise” and “Solutions in Disguise” that I explain in my book “Succeeding with Agile Governance”.
- Symptoms in Disguise. “Symptoms in disguise” occur when a symptom of the problem is mistaken as being the problem itself. If we have flu, then a decongestant medicine will treat a symptom, but it won’t cure the flu itself. Doctors are trained to distinguish between symptoms and diseases. In everyday life and in business, we are often not so discerning. As an example, we might decide that “it is a problem that our sales are lower than expected” and therefore we might think the solution is to increase our sales team’s efforts. But perhaps our poor sales numbers are a symptom of low quality in our product, or are because a competitor has leap-frogged us. In which case we would merely be treating the symptom of low sales by making our sales teams work harder, rather than the actual root causes of low quality or loss of feature competitiveness.
- Solutions in Disguise. “Solutions in disguise” occur when we express the solution to a problem as being the problem itself. In business we see this often: when we complain that “we don’t have enough staff” (implied solution: “hire more people”) or when we argue that “we lack training” (implied solution: “pay for more training”). From these examples, we can see that a “solution in disguise” might in fact be an appropriate solution to the problem – maybe we really do need more staff or more training … but how are we to know that? By framing the problem as a predetermined solution, we have precluded consideration of any other options, and rendered any rational analysis of the underlying root cause of the problem moot.
Understanding the problem, and deriving ‘solutions’ for the problem, is in my experience a surprisingly tricky and iterative process. The reason why the facilitation is so difficult for a 5 Why’s exercise is that there are typically multiple solutions to any stated problem, and what may appear to be a problem at first glance often turns out not to be a problem worth solving, or being merely a solution in disguise, or a symptom in disguise.
5 Why’s is a type of brainstorming, and suffers from many of the drawbacks of that technique – convergence and group-think being two. Unlike brainstorming, 5 Why’s is often more ‘filtered’ (all ideas should be considered in brainstorming, but 5 Why’s encourages each idea to be analyzing before being accepted). Just like in brainstorming, the facilitator should encourage the group to look beyond the obvious. By our very nature we are very quick to drive to ‘solutions’ before the problem is fully understood This often leads to early convergence on a sub-optimal approach. The facilitator therefore is battling against individual’s desire to just ‘solve the problem’, and at the same time the tendency of Group-Think to converge.
Ries presents the idea of ‘proportional investment’ – that you should spend some amount of effort to mitigate each level of cause (if you ask “Why” five times, then there are five levels). It will take time and several iterations to identify exactly what these levels are, given the issues discussed above. The length of time that it takes to do this exercise properly is fairly significant. Done properly the process can take several hours, even for a smallish problem. The process is aided by splitting it into multiple sessions, so that some cleaning and processing of the information can done between sessions.
When facilitating a 5 Why’s session, I have found that it is often best to project a document or spreadsheet where all participants can see it. I will create a tree diagram, or a list with multiple levels of indentations. I aspire to using IBIS (Issue-Based Information Systems) – which is a type of tree diagram used for Wicked Problems, but have yet to get that to work successfully in a meeting.
I do however, like Ries’ suggestion that you tackle as many of the levels of problem as practically possible. Often the root cause, although nice to get to, is expensive and complex to address, and I can see how multiple countermeasures to what are otherwise symptoms of the underlying root cause will make your system robust, and extremely adaptable. The real insight for me in this chapter, though, is to understand the power of doing this type of analysis and proportional investment every time anything goes wrong. Often this type of 5 Why analysis is done incredibly infrequently, if it is done at all, leading to large change-initiatives which deliver results so slowly that the whole method of careful analysis is brought into question. Thanks Eric!
