How to run a prioritisation workshop.
A two-hour session that ends with a ranked list, owners and dates. The trick is simple. Everyone scores in private before anyone argues.
Why most prioritisation meetings go wrong
You put the list on the screen and ask what matters most. The first person to answer sets the anchor. People adjust from there instead of thinking afresh. Anchoring is one of the most reliable findings in decision research, and meetings are full of it.
Then hierarchy takes over. When a call is close, the most senior person decides it, often without meaning to. Quieter people agree, or say nothing. You leave with a list that looks agreed and isn't.
A good workshop separates thinking from talking. People score alone first. The group then talks about the gaps.
Before the workshop
- Name the decisionWrite it as a question. "Which three initiatives do we fund this year?" beats "priorities".
- Collect the options in advanceAsk people to submit items a few days before. You start with a full list instead of a blank wall.
- Clean the listMerge duplicates. Make each item specific and short. Aim for 8 to 15 items. More than 20 and people tire.
- Agree what the axes meanIf you use urgency, set a time frame. If you use importance, name the goals it should support.
- Decide on groupsIf you want to compare views, such as leaders and team, or advisers and operations, set the groups up now.
A two-hour agenda
| Time | Activity | Purpose |
|---|---|---|
| 0:00 to 0:10 | Open: the decision and the rules | Everyone knows what's being decided and how. |
| 0:10 to 0:20 | Add missing items | Anyone can add an option from their phone. |
| 0:20 to 0:35 | Private Eisenhower vote | Each person places every item on urgency and importance. |
| 0:35 to 1:05 | Reveal and discuss the splits | Start with the items people disagree on most. |
| 1:05 to 1:15 | Second vote | See who changed their mind and why. |
| 1:15 to 1:35 | 100-point budget on the shortlist | Force a ranking among the items that survived. |
| 1:35 to 1:50 | Owners and dates | Every chosen item gets one name and a "done when". |
| 1:50 to 2:00 | Confidence vote and close | Check the room believes the plan. Low scores are a warning. |
Running a full strategy day instead? See strategy day agendas for a full day, a half day and a quarterly session.
The exercises
1. Private Eisenhower vote
Each person places each item on a grid of urgency against importance. Nobody sees anyone else's answers until everyone has finished. The result shows each item's average position and the spread of votes. A tight cluster means agreement. A wide cloud means people see it differently. How the matrix works.
2. Discuss the splits, not the averages
Skip the items everyone agrees on. Go straight to the ones with the widest spread. Ask someone from each end to explain their score. Often they're reading the item differently. Sometimes one side knows something the other doesn't. Either way, that's the conversation worth having.
3. Second vote
Vote again after the discussion. If people moved, you've learned something. If they didn't, you have a real disagreement to settle, and you know exactly where it is.
4. 100-point budget
Take the shortlist. Give each person 100 points to spread across it. This breaks ties and stops everything ending up as a top priority. How it compares with other methods.
5. Owners and a confidence vote
Each chosen item gets one owner and a clear "done when". Then everyone rates, privately, how confident they are the team will deliver it. A low score now is cheaper than a missed goal in six months.
Running the room
- Vote before you talk, every time. Including the second round.
- Keep answers anonymous. Show groups if you set them, never names.
- Ask the most senior person to speak last. Agree this with them beforehand.
- Timebox each split. Five minutes per item is plenty. Park anything that needs data you don't have.
- Write decisions as you go. Don't leave it to memory or a photo of a flip chart.
- Close with owners. An item without a name won't happen.
Running it remotely or hybrid
The same agenda works on a video call. Share the host's screen so everyone sees the reveal together. Post the room code in the chat. People vote on their phones or in a second browser tab. Private voting matters even more online, where cameras-off attendees are easy to overlook.
After the workshop
Send the results the same day. A short report with the chart, the decisions and the owners. In Strategy O'Clock, the Export tab gives you a report, a CSV and a chart image before you leave the room. Then put a date in the diary to check progress, ideally at 30 and 90 days.
Run the voting with Strategy O'Clock.
Private votes, the spread on show, decisions captured, report exported. Free, no accounts.