ICE, RICE, WSJF or How to Organize Your Backlog Effectively

Weighted Shortest Job First

Greeting to all who wish to learn how to effectively build a multitasking backlog. My name is Fedor Gvozdev, I am the founder of HolySkin Korean cosmetics online store. I have been working on this project for 8 years and have repeatedly faced difficulties with prioritisation. In this article, I will try to share my experience and present the best models that have helped us out more than once.

Introduction

In the process of planning there always comes a time when a new strategy or reprioritization is required. Many managers face chaotic proposals and opinions that result in unclear goals, loss of motivation and abandonment of important features.

A logical question arises - «How to prioritise the work of a team effectively?»

Effective organisation is the key to success of any project, but how to build it in a competent way, excluding endless calls, disputes and presentations? Backlog work is not as easy as it might seem. In order to achieve real results at minimal cost it is worth resorting to prioritising. There are several mechanisms to do so. They provide an opportunity to avoid verbal skirmishes and resort to quantitative assessments, matrices and charts that will give the team a productive action algorithm.

In this article, we will look into the most popular task prioritisation models and find out why it is important to use them.

ICE is a fast method of prioritisation

This model is one of the easiest tools for goal and objective implementation. If this is your first time using these techniques, this is a great place to start.

The first use of ICE is associated with growth hacking, which is a methodology based on the rapid assimilation of any skills. Most recently this method has become popular in prioritising management.

Let's consider the meaning of the abbreviation:

• Impact is a marker that indicates whether the introduction of an initiative or technology will bring economic growth or an increase in another key indicator against which the tasks are evaluated.
• Confidence is an indicator which shows сonviction in the success of the execution. It is often based on the two remaining criteria.
• Ease is an indicator which assesses the labour and resource intensity of the project.

Evaluation process

The mechanism uses a scale from 1 to 10 for each indicator, then by multiplying the values of the three components, the final ICE Score is obtained. All the features are ranked by importance.

It is very important to coincide the applied scale and your understanding of the position of each indicator. This way the team will have a full understanding of the ranking.

Evaluation example

Collect a list of tasks, and evaluate them on a 10-point scale. Start with the first indicator - impact.

Pros:

• It is easy to understand the algorithm;
• High speed of decision making.

Cons:

• Subjectivity. This model lacks objective assessment and calculation. Relativity of the results can cause loss of an important task or its removal from a higher position.

RICE

The system for evaluating each of the features according to four indicators included in the acronym. The results of this technique are more versatile, and can be used for backlogs where structuring needs more attention.

The components of RICE and ICE are quite similar. Nevertheless, they differ in the level of objectivity and in the assessment process.

Let me explain each indicator:

• Reach shows the number of people affected by the feature or its implementation.
• Impact indicates the benefit from the implemented feature to the final product or the entire project.
• Confidence is an indicator which shows сonviction in the success of the execution. In the RICE model it is measured as a percentage and allows you to correct the situation when there is no concrete evidence of the impact.
• Effort characterises the labour costs. It is expressed in the number of people involved in implementation of one project every month.

Just to clarify: if the project consists of several stages - planning (1 person) - 1 week, design (1 person) - 2 weeks, development (1 person) - 3 weeks, then in total we get 3 team members for 6 weeks of work. In this situation effort equals two.

Evaluation process

Further calculations are reduced to the use of one formula.

Evaluation example

Evaluate each factor according to the method described above, and calculate the total number of points according to the formula.

Results have a wide margin, that is why the choice is obvious. The team agreed with the outcome. Due to a large number of clarifying criteria we were confident in this prioritisation.

Pros:

• Quantification of the indicators reduces the level of prioritisation subjectivity.

Cons:

• The results may need to be reassessed. The results of the RICE model cannot be considered final, since it still remains a more systematized model of the team's confidence in a particular feature.

The model can be considered effective in medium complexity backlog circumstances, but it is important to understand that prioritisation can also be refined.

WSJF – Weighted Shortest Job First

The most versatile and efficient prioritisation model. It considers the optimal number of criteria, which allows you to rationally structure your work.

The name of the model is an abbreviation of the phrase "Weighted Shortest Job First», which basically means that the most important and simple tasks are paramount. This is the main idea of the model. By the means of this model you can obtain a list in which tasks will be ranked by assessing implementation complexity and effectiveness for your project.

Calculations in this model are reduced to one simple formula. The complex part is  hidden in the numerator, since the cost of delay is the sum of three evaluation criteria. It is this component that makes WSJF really effective.

Let's consider all the components of the model:

1. Cost of delay (= User-Business Value + Time Criticality + Risk Reduction or Opportunity Enablement) is the technical complexity of work implementation, which includes:

• User-Business Value is a criterion which evaluates how useful the idea or task will be for your business;
• Time Criticality (temporary or not) shows how important it is to do the task quickly;
• Risk Reduction. While evaluating this parameter, you need to answer the question: “what risks can we protect ourselves from?’’;
• Opportunity Enablement shows the number of potential opportunities.

2. Job size (needed resources) a criterion which includes labour resources, terms of work, costs of freelance work.

Evaluation process

For effective evaluation of features according to WSJF, StoryPoints or ScrumPoints are often used. These indicators show the laboriousness or complexity of backlog tasks. They are based on the Fibonacci number series, which is a numerical sequence where the first number is 1, and the subsequent ones are equal to the sum of the previous two.

The numbers increase non-linearly, making the difference between tasks with 1 and 5 StoryPoints more obvious, which makes selection easier.

Since the method organises the entire backlog at once, the evaluation process differs from previous prioritisation models.

It comes down to creating a matrix with final scores. Because of the matrix the filling sequence is strict:

Evaluation should be implemented sequentially by one column, starting with the most insignificant, which is assigned as 1. Each column must contain at least one 1.

Subsequently, the list of tasks should be ranked based on the obtained results, where the highest WSJF score means the corresponding implementation priority.

Evaluation example

As an example, let's take a backlog with three features from our online store. The task is to evaluate each function by the three Cost of delay criteria. Let me remind you that the assessment is made according to the Fibonacci series.

The Job size is considered as human resource costs. This algorithm was already discussed earlier.

After the assessment we get following table:

In the WSJF model, the highest score is given to a priority task, which is most efficient in terms of time and resources.

In my opinion, this mechanism has an exhaustive number of tools and therefore is most suitable for prioritising any backlog. It deals with both easy and tricky prioritisation tasks. At the moment WSJF is the favourite of our team.

Pros:

• one-time prioritisation of the entire backlog;
• effective rating scale;
• important and relevant criteria.

Cons:

• the possibility of intuitive results if the communication between business and performers is poor.

The WSJF model works successfully and efficiently in most cases. It is suitable for organising both multitasking and primitive backlogs.

Conclusion

Prioritising is a complex and lengthy process. The framework in this case is an excellent assistant, but it still needs to be controlled by the manager.

Even though mechanisms such as RICE, ICE, WSJF do not always exhaust the work on ranking priorities, they are still effective and can change the vector of work quite a lot. These changes can save you a lot of money and rid you of useless tasks.

RICE and ICE are great options for weekly team meetings. They help clarify the task and motivate the team quickly.

WSJF is a more complex tool for multi-tasking backlogs, its effectiveness has been confirmed more than once in my personal experience. The model can be used for projects where possible losses may be too high.

Reasons for using prioritisation models in your work:

• The whole team is involved in the decision-making;
• Saves you a lot of time;
• Rids you of endless calls, presentations and meetings;
• Effective formation of goals by the Smart model.

The choice of a prioritisation model is based on the complexity of the backlog and the goals of the manager. It is always worth trying several mechanisms, then by personal experience you will be able to choose the most effective and accessible model.

Wish you successful projects!