Imagine you are managing an R&D Lab for a large Corporation. The phone rings and you pick it up to be greeted with almost breathless excitement by a senior Marketing Manager. He is fresh from a meeting with his team and has a critical new development that he wants the lab to work on. He has completed the chartering form and he is eager to plan the kick-off meeting. But you know the lab resources are already consumed with the existing project portfolio. To add any new project into the portfolio will require one of three things to occur: add resource, rob resources from other initiatives, or kill one or more of the current projects to free up the needed resource. This is an all-to-common dilemma: What do you do? In my experience the most popular option is the only one that is unacceptable: daylight robbery! Time and time again I have seen great new projects added to an already resource-constrained portfolio without making any adjustments to the portfolio. This leads to a significant diminution of overall performance and, most egregiously, will slow the delivery of the best projects.
Many people have studied this phenomenon (see for example Reinertsen D., “The Principles of Product Development Flow”, Celeritas, 2009) and very often an argument is built around Little’s Law (Little J.D.C., Operations Research Vol 9, pp 1109-16, 1967). This is a very simple relationship, unlike much of what can be found in queuing theory, which effectively states the following: For any given system in steady state, the average queue time is equal to the amount of work in the system divided by the throughput of the system. Thus for your development activities, the average wait time of an initiative is equal to the number of projects the group is working on divided by the average time it takes the group to complete a project. This is intuitively obvious and would say that if you want to get initiatives done more quickly then you should attempt fewer of them! Conversely, the more projects you attempt the fewer you will complete (in a given time).
I have heard it argued that Little’s Law does not apply to development projects since the average time to complete projects is so variable and/or that there are often internal delays (arising from such things as awaiting customer feedback or to get time on a production line for sample production) associated with projects. Whilst both of these things are clearly true, they do not affect the simple and universal truth of Little’s Law: It applies. My advice here is threefold:
Develop a simple scoring mechanism for screening all chartered projects. Many companies create a simple two-by-two matrix with one axis representing some form of value (reward) and the other the probability of success (risk). Projects which score low on both are often correctly referred to as ‘dogs’ (and are to be avoided!) whilst the ‘pearls’ shine bright at other end of the spectrum.
Create a ‘fast lane’ for the best projects (the brightest pearls) i.e. those projects which score the highest (and above an established threshold on reward and risk) during the screening process. A project that enters the fast lane should have a dedicated core team (effectively increasing the throughput) and get expedited attention when waiting may occur.
When attractive new projects are considered for formal resourcing, after that call from the Head of Marketing, and entry into a portfolio, apply Little’s Law. Recognize that deploying already committed resource to a new project will damage the output of the entire portfolio. If all of the needed resource is not available, you must either put the new project on hold, kill another project, or add resource.
In our experience, companies can benefit greatly from insights into Little's Law by using outside expertise to supplement their resources efficiently. At FlexR&D majority of our clients have engaged us to not only come up with deep, cost-effective expertise on a particular subject matter but also as a strategy to protect their entire portfolio against constrained resources.
I'm busy working on my blog posts. Watch this space!