Understanding the Components of Economic Sequencing (WSJF)

Weighted Shortest Job First, or as I like to call it, Economic Sequencing, is a critical tool in properly focusing on the right problem to solve next. Far too often I work with executive leadership teams to discuss their strategic goals, and they give me a list of around ten ‘priorities’. My job is to help them understand that if you have 10 priorities, you have no priority. (The word Priority was never pluralized until around the late 18th century, it’s based on the Latin ‘Prior’, meaning first). Having that one key business initiative, with the other initiatives being clarified as ‘Important, but not the priority’ is critical to limiting Work In Progress (WIP) and delivering real value. Without this clear sequencing based on projected economic outcomes we end up working on too many things at once, and delayed or prevented delivery on the most important thing.  Don Reinertsen’s work (among others) has shown that we get more done in the long run by focusing on one thing at a time.

WSJF in the SAFe® world utilizes proxy values for the Cost of Delay measurements that Rienertsen advocates.  (Note: Outside of the SAFe® world this is typically referred to as CD3, or Cost of Delay Divided by Duration).  These 3 CoD factors, User/Business Value, Time Criticality, and Risk Reduction/Opportunity Enablement, are very often confused or misunderstood, resulting in watered down impact of the process. I’m going to use the analogy of a sail boat to help illustrate these three elements.

Here’s the scenario.  We have a sailboat that we want to sell to a customer, but it has some issues.  The paint is peeling, the motor needs work, and there is a leak in the hull that we have yet to find.  Which problem should we correct first?  Let’s use relative sizing with modified Fibonacci to determine the CoD factors for this effort.

Let’s start with the User/Business Value (UBV) aspect for each effort.  Looking at the boat peeling in the sun, it doesn’t look very pretty, so we probably say that the Paint initiative is our highest issue.  But, what’s our lowest UBV?  Probably the motor, since the motor does run, so let’s assign that our 1.  Looking at the other two items relatively, we then decide that the Leak effort is probably a 3 since it’s causing the boat to settle lower in the water.  Based on how bad the paint looks, we are putting the Paint effort at a 13.  So we now have, in relative UBV ranking, Motor = 1, Leak = 3, Paint = 13.

Time Criticality (TC) is one of the confusing ones for most people, as they tend to think only of deadlines.  The truth is, unless you are building systems that have to align with the lunar cycles or other unchangeable events, deadlines are created by someone, somewhere.  We have to look at time criticality as the loss of potential value over time more than by deadlines, as that is the true impact to ability to deliver the most valuable thing first.  So, where do our 3 initiatives fit in with TC?  Since our motor is running now, and we believe it will continue to run, we are going to put this as our 1.  This does not mean it has no time criticality, only that of our given opportunities it has the least amount of TC.  The paint is peeling, but we are not in the rainy season, so our TC for Paint is relatively low at a 3.  However, the leak is getting worse, and we can’t fix a motor or paint a boat if it’s on the bottom of the ocean, so our TC for the Leak is 13.  The potential loss of value increases greatly the longer we delay.  So, our TC is Motor = 1, Paint = 3, and Leak = 13.

Last comes the Risk Reduction and Opportunity Enablement (RR/OE).  These two elements work well as a combined value since rarely are we able to reduce risk and open new opportunities at the same time.  Risk Reduction increases when we see an opportunity to lower a potential risk to the enterprise or the customer.  In our case, the risk of the Leak is greater than the risk of the Motor failing, but both are relatively high concerns. The Paint has little risk, and while there is an opportunity for increased margin on the sale of the boat, based on greater interest and competition among buyers, it is still our lowest RR/OE value.  Based on this understanding, we are assigning RR/OE as Paint = 1, Motor = 5 and Leak = 13. 

eater than the risk of the Motor failing, but both are relatively high concerns.

Now we can add these proxy values up to get to our forecasted Cost of Delay. 

Paint = UBV (13) + TC(3) + RR/OE(1) = 17

Leak = UBV(3) + TC(13) + RR/OE(13) = 29

Motor = UBV(1) + TC(1) + RR/OE(5) = 7

This tells us that our greatest cost of delay is in not fixing the leak, but it still doesn’t tell us which one we should focus on first.  We have to add in the cost of implementing these initiatives to the equation in the form of Job Size (JS).  This is also a relative number based on our understanding of the complexity of the effort, the uncertainty of the solution, the knowledge (or lack thereof) in the issue, and the over amount of effort and complexity.  Since we are familiar with the process of stripping the paint, prepping the hull, and repainting, but it’s a fair amount of work, we are thinking the Paint would be the largest job size.  The Motor should be a simple fix, but due to our lack of mechanical skills, the Motor has a great deal of complexity, so we are thinking this falls some just below the Paint.  The cause of the leak is readily apparent, and does not seem too difficult, so it’s our lowest job size.   That gives us a Job Size of Leak = 1, Motor = 5 and Paint = 8.

Applying the Weighted Shortest Job Size equation of WSJF = COD (UBV + TC + RROE) / JS results in this ranking:

Leak = CoD(29) / JS(1) = 29

Paint = CoD(17) / JS(8) = 2.125

Motor = CoD(7) / JS(5) = 1.4

This clearly tells us (by a wide margin) that despite how bad the boat looks, the first thing we should focus on is the leak.  It may be hidden from view (eg. Technical debt, architectural or infrastructure needs) but focusing on the Leak first will result in the greatest economic outcome.  This type of data can greatly help with the typical ‘prioritization’ methods that usually result in HiPPO decisions (Highest Paid Persons Opinion), and instead use localized information to create the greatest benefit.