Published May 4, 2022
Did you use stack ranking before?
This is a widely distributed method to get a list of things into order.
It orders items from most valuable to least valuable, revealing the most valuable or most important things. Therefore, it is often used to create rankings and for prioritization.
Stack ranking is being used in many different areas in a variety of use cases. Typical use cases involve ordering goals, tasks, ideas, or feature requests on a roadmap.
There are several approaches to performing stack ranking for a list of items. Although the methodology for the different approaches can slightly differ, most of them are based on calculating a score for each item that should be ranked. The final order of elements corresponds to the score that each item gained during the ranking.
How is the score calculated? Let’s have a look at how the score is calculated in two commonly used variants: criteria-based and pairwise stack ranking.
What are the steps that you do in both of these variants and how do they differ from each other?
In criteria-based stack ranking, you start by defining scoring criteria. Let’s assume you have a list of tasks that you want to prioritize. Your scoring criteria could be the value of the task and the duration of the task (2 criteria). It is important that you are able to express the criteria value for a task as a numerical value because you will need to calculate the score from the values that you assign. The value of the task could be how much money you earn from performing the task or much time it helps you to save in the future. The duration can be the number of work hours or workdays you need.
You calculate the score of an item by assessing the criteria values for each item and then applying a weighted formula that calculates the score for you.
This variant works differently. The approach requires fewer separate steps and can therefore be simpler to perform. It does not use any criteria (in the simplest variant). Instead of defining criteria and evaluating criteria, you calculate the score of an item by comparing that item to each other item on your list. This approach is often also referred to as ‘Pairwise comparison’ or ‘paired comparison’. Each time an item is more important than the item you compare it to, the item gets a scoring point.
After comparing all items with each other, all the scoring points of items need to be counted. The sum of the scoring points represents the rank of an item in the stack ranking.
Many prioritization techniques rely on the existence of additional data about the things to prioritize. While it is usually not difficult to collect additional data about a single item, the amount of time can significantly increase if you have a long list of items to rank. Also, many times the data you will try to collect about the items can be ambiguous: if you work in a team of multiple people and try to assess or estimate data about goals or tasks, you can have very different opinions. That situation might prevent you from quickly reaching a consensus without spending a lot of time in meetings, delaying the whole prioritization process.
A great advantage of stack ranking without criteria is that it forces you to make a real choice about the priority of an item. Therefore it is also sometimes called forced ranking. This characteristic is great because many times teams suffer from ambiguous priorities for the reason that every task or goal is treated as highly important. In forced ranking, most of the time the final score that each item has is unique, which means that you also get a clearly ordered list of priorities with unique ranks. The only exception is if you have very few items, then the scores of items might sometimes be similar.
Having similar scores makes your prioritization less expressive, but does not lower the quality of your result: you still have the information, which of two items is the more important one when you look at the result of a direct comparison, and you can use that information to make a more confident decision even if two items share the same final rank.
We created spreadsheet templates for you to help you getting started with stack ranking. Each template contains summarized instructions on how to use it inside of the template. You can change the options names and customize it as you need.
Please choose between the two types of methods: a) pairwise comparison (without criteria) and b) with criteria.
To get a copy of the Excel template, just click on the link and download the Excel sheet to your drive.
To get a copy of the Google Sheets template, click on the link. It will open the sheet online. A linked spreadsheet is read-only. After opening, you can download the template or create a copy. To create a copy, use the menu above the table. Click on ‘File’, and then click on ‘Make a copy’.
The instructions for the sheets for pairwise ranking are included inside the template.
This template includes scoring criteria to assess items.
You can use the template with multiple people: one of the criteria can be assessed by multiple persons. Each person participating can vote the impact of an item (you can change the name of the criterion). The final impact value used will be the average impact of everyone who voted. Other criteria in this template (such as effort) is meant to be entered by one person only because it assumes that you have a common understanding of the effort. However, you a free to choose which criteria you want to vote with multiple people and which one not.
Pairwise comparison, Excel template (xlsx format): Download template Excel
Pairwise comparison, Google Sheets template: Download template Google Sheets
Weighted (with Criteria, multiple people): Download template Google Sheets