There are many blogs and adaptations and various write-ups about the Fruit Salad Game when it comes down to Scrum Games. However, as useful as they are I feel that they were a little limited and biased. Here I’ve adapted the classic and from experience expanded it to cover more, in a way that is realistic to many delivery teams.
- Part 1: Problem definition
- Part 2: Solution design
- Part 3: Prioritisation
- Part 4: Backlog grooming
- Part 5: Estimation
- Part 6: Sprint Planning
- Part 7: Sprint Simulation
- Part 8: Retro
To do all these activities, you will need the following:
Time: 4 hours but you can split them up in several shorter workshops with your team.
Resources: Pens, paper, whiteboard.
People: 2 people or more (game facilitator and participant(s)).
They will help your team to better understand the various agile-scrum practices and how to reach consensus in defining, estimating and delivering work regardless of the differences in experience and skill-set amongst them.
So let’s start with the first part in this article – Problem Definition!
Action: Define the problem and set the scene.
As facilitator (and party host) begin with explaining that we’re throwing a party tonight and that we have everything prepared except a fruit salad. This is our problem definition. For the purpose of this game assume research is complete. Further business requirements, constraints and assumptions are mentioned.
- The fruit salad should be made fresh.
- The fruit salad should be exotic and tropical.
- There are no limitations to how many types of fruit, and how many of each type of fruit go into the salad.
- The fruit salad must contain mango and pomegranate.
- The party starts in an hour.
- We have a magical pantry with every fruit in the world, which never runs out.
- There are 10 guests invited
- We have hired one trainee chef, only they are allowed to prepare fruit.
Action: Allow for questions
The facilitator will now take questions from the team, allowing them a chance to address any ambiguity and get further clarity to prepare for the task at hand.
Some questions you may get:
Q: How many papaya, is just one ok?
Q: What do you mean by exotic and tropical?
A: I mean containing fruits that aren’t available at your local grocery store, and ideally those that thrive in the tropics.
Q: How big does the fruit salad need to be?
A: You tell me, would you be satisfied with one grape as your fruit salad?
Once the team is satisfied with the business requirements and understands with a level of clarity what they need to do, we can move onto part 2: Feature Definition.
Action: Explain the task at hand, for feature definition.
Now that the team understands what they need to build, they need to collectively decide on how they’re going to do it.
Action: Ask the team to come up with a list of fruits (features)
Ask the team to come up with a list of 10-15 fruits that may be candidates for the fruit salad.
E.g. Say they come up with a list like this:
Just make a note and remind them if they haven’t included the 2 mandatory fruits mango and pomegranate.
As a conclusion for these 2 first exercises, your team will now understand how important collaboration is and how they can come up with very different options with the same initial requirements.
Once the team is satisfied with the list of fruits/features for your salad/product, we can move onto part 3: Prioritisation.
It’s prioritisation time!
We all know priorities change and there’s a number of factors which lead to their change or to the exercise of prioritisation to be a difficult one. To further add to our requirements we’ve also listed additional feedback gathered that may or may not influence prioritisation
Provide the group with a list like this:
- Everyone at the party loves tangy citrus
- Half the people like (but not love) hard fruits
- Large, complex fruit is always appreciated and favored.
- Primary colours make the fruit salad appealing and enhance the overall appearance.
- Sweet fruits are a big hit!
Based on this information and the earlier constraints. Ask the team to prioritise what they would like in their fruit salad based on what they now know. throughout the process they can continue to ask the facilitator (product manager) questions if they need clarity. Remember we already know that the fruit salad must contain grapes and mangoes.
Here’s a sample prioritised list (1 being the highest priority):
Observe and note the discussion and thinking process during the prioritisation. If not clear ask them to talk out loud and discuss, and come to a mutual agreement. Facilitate to ensure that this moves on swiftly. What was the method? Why was grape at the top, why was jackfruit at the bottom?
Backlog grooming time!
Now that we know the features and the order of priority, let’s dive into the details.
Create a card for each of the fruit, and on the back, as a group, write the acceptance criteria that would qualify as this card or fruit being successfully implemented into the fruit salad. I.e. ‘done’.
At this time, the team is allowed again to ask as many questions as they like. An opportunity for the facilitator to have some fun too and introduce a few ‘spanners’ in the works.
E.g. Banana card → Team decides the criteria is:
- Peel banana
- Slice banana multiple times
- Add slices to bowl
Facilitator Spanner: We’ve just had a work from our UX team that there is a pleasant experience increase of 40% by thinly slicing and quartering the banana. We quarter and thinly slice the bananas if they are to go in.
So.. reluctantly, the team will likely refine the 2nd acceptance criteria to ‘Thinly slice the banana’ and add a 4th criteria of ‘quarter the slices’.
Next fruit: Mango → Team decides the criteria is:
- Wash the mango
- Cut the mango
- Add it to the bowl
Facilitator Spanner: We’ve just had a work from our UX team that there is a pleasant experience increase of 20% by having a mango 2 ways in a fruit salad. I’d like pulp from the half of it, square cubes from the other half.
Team will need to add in 2 more steps:
- Squeeze out pulp
- Cut cube shapes
You can continue throwing spanners for the rest of the fruit, or simply stop at these two and let the team go ahead with the rest. Coconut can be particularly difficult, note how the different ideas and skill sets come through within the team when it comes to breaking open, slicing out the contents and adding the coconut blocks to the salad.
It’s time for some estimation.
Relative or Individual – You choose.
So it’s up to you as the facilitator how you want to do this. You can start at the top of the priority list, e.g. Grape and ask the team to estimate one by one down the list (individual), or you can first organise the list into a list of higher/lower complexity and then do the estimation (relative).
However, it’s always fun for us planning poker cards. So let’s take the opportunity to do so!
- Hand out cards to each person.
- Have the teams announce the first fruit, e.g. grape and read the assumption criteria.
- Ask each person to take a number from their deck (Fibonacci numbers) and place it upside down in front of them.
- On the count of 3, everyone turns over their cards.
- The persons with the highest and lowest numbers explain why they gave that estimation, and we discuss.
- Repeat the estimation for grape and once everyone is in agreement, move onto the next fruit. Generally, you’ll see a 1 or a 2 at most for the grape. Both are fine. If you start with an 8 for the grape, you’re going to struggle when you get to coconut.
Grape → 2
Mango → 8
Watermelon → 13
Papaya → 8
Pineapple → 13
Apple → 5
Pomegranate → 5
Coconut → 13
Avocado → 8
Banana → 3
Orange → 5
Strawberry → 2
Pear → 5
Jackfruit → 13
Durian → 21
So now we know what the features are and roughly how complex they are to include.
Let’s take a look back at some of the facts and constraints we have to work with.
1 hour to make the salad
As we don’t have the velocity to work off, we don’t know how much our chefs can do in an hour. We need to make some assumptions and put some time constraints in place for this to work.
Let’s begin at the top: Roughly how long should it take to prepare the grape?
Say 10 seconds? If people argue, act it out. Grab an invisible grape, pretend to wash it, cut it and add it to the bowl. Don’t get too stuck on this, go with an estimate quickly and move on.
Let’s take an apple, wash it, cut it, add it. If we look at our estimation, it’s also significantly less in estimation points and therefore should take less time to deliver. So, say 30 seconds
And so on… continue until you’ve given a rough time to prep each fruit.
Sample time list in relation to estimation:
Grape → 2 → 10s
Mango → 8 → 60s
Watermelon → 13 → 90s
Papaya → 8 → 60s
Pineapple → 13 → 90s
Apple → 5 → 40s
Pomegranate → 5 → 40s
Coconut → 13 → 90s
Avocado → 8 → 60s
Banana → 3 → 20s
Orange → 8 → 60s
Strawberry → 2 → 10s
Pear → 5 → 40s
Jackfruit → 13 → 90s
Durian → 21 → 150s
So here’s where it gets serious.. And mathematical!
We have 60 minutes i.e. 3600s.
There are 10 guests arriving, meaning we need to make enough per plate to feed 10.
Ask the team what they would add?
Here’s the time list when preparing for 10:
Grape → 2 → 100s
Mango → 8 → 600s
Watermelon → 13 → 90s (1 is enough for 10 dishes)
Papaya → 8 → 60s (1 is enough for 10 dishes)
Pineapple → 13 → 90s (1 is enough for 10 dishes)
Apple → 5 → 400s
Pomegranate → 5 → 400s
Coconut → 13 → 90s (1 is enough for 10 dishes)
Avocado → 8 → 600s
Banana → 3 → 200s
Orange → 8 → 600s
Strawberry → 2 → 100s
Pear → 5 → 400s
Jackfruit → 13 → 90s (1 is enough for 10 dishes)
Durian → 21 → 150s (1 is enough for 10 dishes)
E.g. team takes:
- 10 grapes
- 10 mangoes
- 10 Pears
- 10 Strawberries
- 1 Pineapple
- 10 Apples
Will we make it in time?
100+ 600+ 400+ 100+ 400+ 90 = 1690s leaving us with just 1910s
Do we pack more in? Are we happy with our product? Do we foresee any risks or delays?
Let the team decide, and tell you when to hit the start sprint button.
Sprint started – Its on!
Chef is working away in the kitchen. But like Mike Tyson said.. “Everybody’s got a plan, until they get punched in the face”
Here come the spanners!
SPANNER: We bought a pack of seedless grapes but one of the grapes turns out to have a seed, chef has a seed fobia it, and passes out, we lose 10 mins while he regains consciousness. I.e. 600s of our 1910s. Leaving us with 1310s.
SPANNER: I Just got a call from 2 guests, and they’re bringing their partners!!! That’s a 20% guest increase, adding 338 seconds if we want to make two more bowls. Leaving us with 972s.
SPANNER: Oh dear…. Whilst slicing through that massive pineapple. Chef sliced a finger and needs medical attention. It will cost us a further 10 minutes to provide assistance. Which is 600s… we only have 372.
SPANNER: The knife has gone blunt! We’ve gone out to buy a new one and that’s another 10 minutes. Which is 600s…… guess what.
We’re not going to make it in time!
What do we do!?! We’re not going to make it in time, argh… Panic panic panic!
It’s time to reflect. What would we have done differently? How can we prepare better for the next party? What should we keep doing, stop doing and start doing?
At this point we can end the simulation, all is okay. Ask the group, under this stress what is the best approach. How would they deal with the situation? What would they pull out of the fruit salad or what would they replace it with? Would they call off the party or would they ensure that we deliver something, even if its less than what we intended, in order to get some value (salad) out to the customer? Discuss.
The Fruit Salad Game is a simple way to learn the agile processes with your team. It lets them practice methodologies within an easy and