WSJF stands for Weighted Shortest Job First. It’s a technique used in scaled Agile framework (SAFe) to prioritise jobs—epics, features and capabilities—according to their value relative to the cost to perform it. Basically it’s a way of ranking a list of features in order to maximise the outcome—the value produced—with a constrained capacity to produce it. The job with the highest WSJF (value over the cost) is selected first for implementation.

It’s not a brand new idea, long ago a very similar technique known as Value Engineering (VE) representing the ratio of function to the cost was used.

Value engineering (VE) is a systematic method to improve the “value” of goods or products and services by using an examination of function. Value, as defined, is the ratio of function to cost. Value can therefore be manipulated by either improving the function or reducing the cost.

Computing the WSJF

Estimating the value

In SAFe we are talking about the Cost of Delay (CoD). 3 elements contribute to it that can be estimated using the modified Fibonacci numbers (1,2,3,5,8,13,20)

I see several disadvantages with this method:

Estimating the duration

Instead of using duration which is less straightforward to compute, the job size can be used—even if team capacity is not the same, duration is proportional to the size. Since size is always estimated in story points it can be used as is.

Computing the ratio

Now the simplest part of the method putting all together and sort the result. For fun I’ve made the example with some Python code.

import random
from random import sample
import pandas as pd
from tabulate import tabulate


# Modified Fibonacci sequence
fib = [1, 2, 3, 5, 8, 13, 20]

# Generating sample data for the Cost of Delay
df = pd.DataFrame({'feature': ['A', 'B', 'C'],
                   'b_value': sample(fib, 3), 
                   'risk': sample(fib, 3), 
                   'time': sample(fib, 3)})

# Computing the Cost of Delay (CoD)
df['cod'] = df['b_value'] + df['risk'] + df['time'] 

# Generation sample data for the duration
df['duration'] = sample(fib, 3)

# Computing the Weight
df['weight'] = df['cod'] / df['duration']

# Implementing the featuer with the higher weight
df.sort_values('weight', ascending=False, inplace=True)

print(tabulate(df, headers='keys', showindex=False))

And the result is that you should implement the feature “A” first!

feature      b_value    risk    time    cod    duration    weight
---------  ---------  ------  ------  -----  ----------  --------
A                  2       2       5      9           1      9
B                  8       3      13     24           8      3
C                 13       8       8     29          20      1.45

References / Further reading