We have just released a new version of our Planning Poker application for the iPhone, iPod touch and iPad. It wont cost you a penny as it is free.
Planning Poker is an estimation tool used in the Agile methodology. Rather than asking a single developer to work out how long a piece of work will take, the entire development team decides. And the way this is done is for each person to say how long they think it will take by playing a card. The numbers on the cards vary. And several sets are in use. At Unboxed Consulting we use the Fibonacci sequence. Which gives us the cards: 0, 1, 2, 3, 5, 8, 13, 20, 40, 100, ? and break. The reason for using such a sequence is that as the estimate gets larger it is not possible or necessary to be that accurate.
The estimate is given relative to a base story, and the number (if using a number system rather than t-shirt sizes) is not an actual time but just the number of points assigned by the team. So if you gve the baseline story 5 points, and the next story was about twice as long you would give it 10 points. With the Fibonacci sequence we can’t go for twice as long. There is no card 10 for. So instead we have to think whether it is less than twice the effort, in which case we vote 8, or slightly more than twice the effort we vote 13.
If everyone votes the same for a story then everything is rosy and you just assign that points estimate to your story. If there is a difference in opinion then anyone who voted at the extremes of the range is given a chance to explain why they gave it that vote. So say we had the votes one 2, three 5 and one 13 then the people who voted 2 and 13 would explain why they did. After their explanation everyone votes again. And you keep doing this until you get agreement. At Unboxed when the difference is small we do the following: Say we get one 2, three 3 and a 5 then rather than spend extra time trying to get agreement we would take the most popular vote. In this case 3. There is no point spending too much time. With the lower scores you could easily burn that number of points trying to get a more accurate estimate.
The first time you do this you will have points assigned to each story. But you will not know how long it will take. So the team has to agree on how many points it can do in the first iteration, and the product owner will prioritise the stories to get the most important ones to them done first. After the first few iterations you will start to get an idea how many points you can do with a certain team size, your so called velocity, so you will be more accurate working out how many stories you can do in an iteration.
Time is something you need to watch out for. Developers on the whole love trying to come up with the entire solution. It’s an estimate. So you need to stop this happening. Someone who knows the story should explain what is required. This will probably be the product owner. Then you can spend a few more minutes with the whole team discussing it in a bit more detail and then you should vote. To help constrain time the latest version of Planning Poker for the iPhone has a timer that runs in the background.
The card sets usually come with a question mark which should be used when you have no idea what the estimate entails, and often there is some sort of break card which you can use to indicate that you think you have been estimating for a long time and need some refreshment or some other type of break, you can play this card.
This version also allows you to define a custom card set if you use a number sequence we don’t support. The custom set cards are not just restricted to numbers. So you could use it if you prefer t-shirt sizes. To get the cup picture for a break you need to enter “break?” into the card. In the actual card set it will be displayed as the cup.
Our Planning Poker application wont make your estimates more accurate. But we hope it makes the process a bit easier and more fun.
If you haven’t got an iPhone, or just prefer a set of physical cards we can help there too. Contact us if you want more details.
The key thing to remember:
It’s an Estimate