Precedence: a prioritisation tool

John Sunart
Oct 23, 2018 · 6 min read

This is a practical guide to using precedence in a workshop environment.


In 2018 I worked with one of the big five energy companies restructuring their internal database systems. In 2016 I worked with the Scottish Ambulance Service defining the criteria by which they’d assess and purchase new equipment. In 2015 I worked with a consortium of hospitality businesses in the north of Scotland to build a coherent tourism offering.

In all of these instances, building consensus from a diverse group of stakeholders was essential. But conversation and groupthink wouldn’t deliver us anything new or concrete.

In the case of the energy company, it was essential that we had a strict hierarchy to produce a development roadmap. For the ambulance service, we needed a weighted scoring system which a third party could use to objectively assess hardware. For the hospitality business, we needed concrete evidence of an auditable decision-making process to apply for funding.

In each of these instances, I used the same tool, which I learned from Norman McNally, who I was lucky enough to work with after graduating. He calls the tool Precedence, and though he learned it in a similar form to this “sometime in the 70s”. The origins are lost in the mists of time!


Precedence is about getting a group to make a rigorous, justifiable decision. It’s most useful for ranking lists, so it might be used for areas for the project to focus on, a list of features, or for building a hierarchy of principals etc. The key value of precedence is it is a decisive, group decision which is a difficult thing to gather usually. It’s also fantastic for building a weighted list of requirements.


For this tool you need a dry run, factor in 35 mins for that, and then 40 mins for a standard 6 item list. For every extra item on your list you need to increase the time by about 5 mins. Don’t go over 10 items, or your participants will be after your blood. Having said that it is a very popular tool after the initial 20 mins of confusion. People feel they are making real, measurable decisions (which they are!)


Gather your thoughts/ideas to be compared (could be generated by brainstorming, discussion and note taking, investigations… Reduce the number through mutual consent (dotocracy or some such method) to around 6–10 options. Also set aside options which are simply essential (In an ambulance everything has to able to be sterilised) and knocking out anything which everyone agrees is unimportant.

Make a grid with the same number of rows (+1) and columns as you have items to be ranked or use post-its like the photo. Write the items you want to rank down the rows, and then the same items across the top (Keep them in the same order, and your handwriting legible or things will get messy) — Don’t underestimate how long this takes, best to do it with your participants getting coffee etc

Grey out the boxes where one item would be compared against itself.

There is going to be one vote for every box above the grey’d out boxes. You’re going to do the maths to fill in the bottom part.

Have another facilitator — or a participant who’s willing to count hands in the air.

Finally count your number of participants, write it on a post-it and stick it next to your precedence table, you’re going to forget it at some point and a reminder is really useful (It’s really good to have an odd number of participants if you can fudge that somehow)

Prep the participants, tell them you’re going to call out two features, and they are going to put their hands up if they think the first feature is more important/relevant/doable/worthwhile…. This is where the confusion will come, and it’s why you do a dry run.

Ok, good to go!

  • You’re going to call out something along the lines of “Item one over item two”
  • Folk who feel the item one is more important raise their hand.
  • Your counter is then going quickly to shout out the number of people who have put their hands up. — Speed is key, or you will be here all day. This is a personal decision, not a group decision. Don’t allow conversation to develop (make sure this is clear, you’re going to have your back turned and you’re going to be concentrating for the next 25 mins)
  • Write that number in the correct box (which is above the grey’d out boxes)
  • Subtract the number of raised hands from the number of participants and put the result in the corresponding box below the grey’d out boxes.
  • Move across the row (Item one over item three..), repeating steps 6–9.
  • It’s time for row two. Keep going. It’s best to try and get through this is one sitting. People will start to help you, and as the pace picks up you’ll get a good vibe from the room as they start to figure out what they’re creating.

When you’re done, and every box is filled, total your rows, and write this in the right hand column. Don’t rush this stage, it’s important. If you feel tired (which is totally understandable) have someone else do this final bit of maths. There is always a mental arithmetic school champion circa 1983 in the room who’s up for showing off.

Now it’s as simple as ranking the totals, the highest scorer is 1, down to the lowest.

If you have two that are the same score then go and compare how they were ranked against each other, the one that scored higher gets a marker to indicate it is above the other (I usually use a +) If you have an odd number of participants this will solve your problem, if you have an even number then this will hopefully solve your problem… If you have three which are the same run a mini precedence with just those items.

Well done. Break, Tea, Cake.

The score you generated for each item can act as a weighting system for analysing a product or service later on in a Weighted Analysis.

Dry Run: Before running this tool in earnest do a dry run with the group. This will get you around the standard problems (Which one am I voting for again…) and also get you into the mental arithmetic game. It should also demo what the tool aims to do. In my experience, the best dry run topic is thumb tacks (pens can work too). It’s also best if you bring props, so collect a few types of thumbtacks together.

  1. Explain a context (Thumb taks for an office environment, for a home environment…)
  2. Brainstorm features of thumbtacks which are essential. Hand out thumbtacks and have people try them to figure out what “features” might mean.
  3. Get down to six, using dotocracy/discussion/common sense.
  4. Do the above process.
  5. What usually comes out on top, but what no one expects to come out on top, is that the most important feature of a thumbtack is the ability to remove it from a wall easily. You can put this dry run to the side and come back to it for a weighted concept comparison if you’re going on to that.

Bonus trick — Changing numbers: If you end up having people drop in and out of a session and you absolutely have to do a ranking for a series of items you can use a fraction to indicate the number of people who participated in each group. The maths at the end is a bit nastier though, certainly not mental arithmetic for most of us.


Thoughts, case studies, and design tools built as I work to…


Thoughts, case studies, and design tools built as I work to embed codesign and Service Design into agencies.

John Sunart

Written by

Facilitator and designer with intent, writing from Brighton, England. Previously a Flitcroft


Thoughts, case studies, and design tools built as I work to embed codesign and Service Design into agencies.