Month: May 2021

2021/05/18

Choosing the Right Opportunities

A critical decision every organization needs to make is choosing the work we will do. Choosing one work item over another is in many ways similar to other investment decisions. As in other investments, we cannot pursue all available opportunities as we do not have enough time, capital or people to divert to each one of them. Hence we need to prioritize the investment opportunities knowing that choosing some means saying no to all the other.

A typical prioritization system will stack a series of options, the things on which we could work, ranked by its potential to generate the value we want as an organization. Assessing the priority we should give to each option is not a trivial task, and many different techniques have been used, typically involving the definition of a scoring card balancing aspects such as organizational strategic alignment, potential financial value measured (for example net present value, NPV, or return on investment, ROI) and risk profile, etcetera.

Donald Reinertsen has described the benefits of using the weighted shortest job first (WSJF), weighting by cost of delay (COD). This prioritization method has amongst many other advantages its dynamic approach. It is based on comparing the cost of not taking each option now, hence delaying it by a unit of time and looking at how much will we lose. More formally, it is the partial derivative of the total expected value with respect to time. One more useful way to look at it is thinking of the total value of the option as a water tank that has a leak and hence the total amount of water in the tank (total value) decreases over time. The longer the time we wait the less water in the tank. How fast it decreases depends on the size of the leak, which is the COD.

This dynamic aspect of cost of delay is what, in my opinion, gives this method an enormous advantage over other simpler approaches, like looking statically at the total ROI. On the negative side is the difficulty of estimating COD, but I must say I did not find it normally harder than estimating NPV or ROI. Learning the art and craft of estimating COD and hence prioritizing our portfolio is a long, arduous but absolutely worthy endeavor. It is learning to see how value evolves over time.

This post is part of a series on planning. In the same series, you can also read:


2021/05/17

The Timing of Planning

Considering the amount of uncertainty increases the longer into the future we try to see, planning once and then executing following our plan does not look like a winning strategy, no matter how much effort we put into planning. We may even be hurting the quality of our planning by doing too much of it at the onset.

What is a planner to do, then? We can spend just enough effort at the beginning planning our endeavor. Then start executing, and refine our understanding with periodic review of our progress and again spending just enough effort to (re)plan our work ahead. “Just enough” is a problematic expression.

How will we know how much is just enough initial planning? I don’t have a satisfying answer yet, but practice has led me to some heuristics I keep refining: If you are still uncovering risks and opportunities that you did not see before, that have the potential to alter your chances of success and which outcomes can be influenced by mitigation or preparedness of any kind, you ought to keep planning. When the risks or opportunities you uncovered had been trivial enough, then it is time to stop. If you have started running in circles, chasing “unknowables”, or pretending you know things you cannot know, it is also time to stop planning. Start with a not so large effort and then progress your initial planning in short loops, when the improvement done over previous cycle does not feel significant, and we still feel we put enough effort and made wise choices, it is time to stop.

Another difficult decision is how often should we review and replan. Again some rules of thumb may help us decide. The farther ahead we can see the longer we can keep our heads down busy with execution without planning again, and more effort we can put in each round of planning. In risky circumstances, unknown terrain or in the presence of unpredictable events we will be wise to plan often, looking ahead to a shorter time horizon, and spending less time doing so for each loop. Each loop should be firmly grounded in reality, considering all feedback we can get from our execution and the events unfolding around us.

No standard planning approach is universally adequate. The answers to when, for how long and how often to plan are highly contextual, and you ought to make a conscious choice. We are all called to lead our organizations making a explicit decision that is grounded in our context and then constantly refinining our approach to this timing questions as get more and more feedback from reality.

This post is part of a series on planning. In the same series, you can also read:

2021/05/16

The Outcome of Planning

“To plan” is much more important than “the plan”. The outcome of planning is not a document, but the things we learned about our current state, goals and impediments; the preparations we made for a possible future in which some uncertain events happen; and the actions we took to get ready for action.

When, after ardous preparations, our plans look like long to-do lists, with the expected duration attached to each task, and an assumption that the future will unfold as the unimpeded delivery of item after item in perfect succession, we are just fooling ourselves and reducing our chances of success.

When we skip planning altogether because the future is uncertain, and we abandon any expectation of being prepared for risks we can foresee but not predict, we are choosing to reduce our chances of success too.

The ultimate aim of planning is increasing our chances of success with a relatively small effort investment. Abandoning after planning is also a form of success, as it costed us little to avoid a likely failure. Avoiding impossible endeavors is only one of the flavors in which successful abandoning comes. Another often overlooked flavor is to complete something doable, but which benefit is not worth the effort. Beware.

2021/05/09

Planning, Forecasting and Committing

Plans are not commitments. If I can commit to a plan, it is a very bad one. If I want to commit to a plan I created before starting, I am accepting I will not learn anything useful during the journey, that is, by actually doing the work. I am not yet ready to accept that, are you?

A plan is the result of planning. Planning, the act, is the important thing. When planning we often detail what it would look like when we achieve what we want, what needs done, the methods by which we can make it happen, the risks and opportunities involved, mitigation and possible courses of action to respond to those foreseeable circumstances. A plan is not an end date, a work breakdown or a rigid course of action, but it can contain some of those things and even multiple variants of them depending on factors unknown or unknowable before we start.

Reassessing our plan regularly as we go is a wise thing to do, even more so when we cannot reach our destination using a well-known, repeatable, and predictable path. We can call that approach progressive or adaptive planning. If nothing new is discovered since last assessment we just keep course with the previous version of the plan, only a bit further along the journey. Often we will need to adapt our plan to accommodate some discoveries as reality unfolds. Possibly, we discover something so different to what we expected than the previous plan does not make sense anymore and we create a new one from scratch. In some cases we may discover our objective is not reachable or it is no longer desirable and we abandon the objective altogether. In short, plans change along the way.

Then, what is it that I can commit to? I believe the answer is more nuanced that never committing or commit to any plan that you create. Business requires some predictability in due dates, quality we can trust or actual cost close enough to our estimates. In order to take a responsible approach to committing, we need to forecast. We use the hard won knowledge we acquired through planning to forecast a set (not one) of foreseeable end results for that variable and their associated uncertainties.

To commit I have to answer a very difficult question: how likely is it that I deliver on the expected result or better? The context will also help me understand how much risk is acceptable in my commitment. Often we have to commit to something under most or all currently foreseeable circumstances. That is industrial-strength likelihood. Hence, the most likely result is not the answer. Instead we should commit to what we can deliver in most foreseeable scenarios, even in the bad ones. Most of the times that does not include unprecedented earthquakes, unforeseen disasters or great new technologies that will reduce the required effort to a tenth, but the ones we can expect. Of course, if I am planning in times of COVID-19, that is at the time of writing this, and I am not including contingencies for outbreaks and other disruptions created by the pandemic, I am being foolish.

When the unexpect-able happens, the black swan, our commitment will not hold. But that is a different story.