Why the Whole Team should Participate in Planning Poker Estimation SessionIs whole team participation (including irrelevant skilled members) fruitful during Planning Poker estimation? It's a big "YES" and this article proves why.
There is a common perception that those team members who will be part of the project should be the ones involved in its estimation, rather than doing estimation with a whole or different non-contributing group.
But the situation is a bit complex in the case of agile teams when they conduct product backlog items estimation. It is because they usually don't know who is going to work on the items. Mostly the teams make that decision during daily standups or while doing Sprint (iteration) planning. This implies that the whole agile team should be part of the product backlog estimation session. But here's the confusing part. Why someone whose skills are irrelevant or not required to deliver the product backlog item should be the part of the estimation session?
Well, there are many reasons that reflect that estimation becomes a lot more efficient if the whole team no matter the skillsets is contributing to the estimation session. This article will list down those reasons and clear your concepts with the help of examples. But first, let's have a brief overview about Planning Poker – the most widely used technique to estimate product backlog items.
Planning Poker is a collaborative estimation technique based on consensus. It begins when a team sits together and the stakeholder or product owner briefs the team about the product backlog item to be estimated. Once the brief is completed, the team is allowed to ask questions and do a discussion around it to clear their concept around the item.
During the session, each team member holds a few poker-style cards that contain the values or points that they will use to estimate. While there is no restriction on which values should be assigned, it is discouraged to opt for precise values. For instance, giving one item a value of 49 and the other one 50 makes it impossible to differentiate the efforts required to complete that item. Therefore, most teams usually use the modified Fibonacci sequence (1, 2, 3, 5, 8, 13, 20, 40, and 100) to give estimates.
The discussion around the item continues until all the members have understand the product backlog item to be estimated. Next, every member picks a card based on what he/she thinks should be the estimated value for that particular item. Afterward, they reveal their cards all at once. If all cards show the same value, then that is declared the team's estimate. However, if there is a big variation, then the ones giving extreme low or high values are given a chance to clear their viewpoints. Afterward, a re-estimate is carried out to reach a consensus.
If you want to learn more about the Planning Poker estimation technique, click here to read a comprehensive article on it.
During estimating product backlog items using the Planning Poker estimation technique, the team members do not reach a consensus based on the estimate that receives the highest votes. It is because Planning Poker does not work like that and gives credits to estimators. Let us understand it with the help of an example.
Consider that a team is estimating a product backlog item where a code written by one programmer is required to be enhanced. So, when a team estimates the item, they will give more credit to that programmer's estimate compared to someone else who wasn't part of the work previously.
In short, all members of the team can contribute to the discussion, ask questions, and give estimates, but the team as a whole will give more value to those members who have more experience working on a similar task in the past.
One of the cool aspects of the Planning Poker estimation technique is that estimates are made relative instead of absolute. What it means is that the team will estimate the required effort to complete the item without estimating the number of hours required to complete it. So, they don't have to dig deeper to allocate the number of hours they want to assign to that item. They just have to understand the item, do some discussion, and then provide the estimate based on their past experience with a similar item.
Due to relative estimation, there is no harm if the whole team participates in the estimation session because they are not estimating the actual hours required to complete the item. Similarly, as the weightage is given more to the relevant skilled members, therefore whole team participation casts no harmful effects on the accuracy of estimation.
When we say that the whole team should participate in the estimation session, then it does not mean that all the members should give an estimate. Although doing a relative estimate is a lot easier than an absolute estimate, but there can be some product backlog items in which not all members are skilled to estimate rightly.
It also does not make sense to ask irrelevant skilled members to give estimates on some items they have zero experience with. But the beauty of the Planning Poker estimation technique is that it lets irrelevant skilled members contribute during a discussion. They can be actively involved in the discussion session, asking questions about the targeted item, highlighting missed spots, and doing similar other contributions.
In short, when the whole team actively participates in the Planning Poker estimation session, the discussion part gets a lot more efficient and so does the team's estimate.
Other than the above-listed reasons, there is another main reason why the whole team should be part of the estimation session, i.e., it improves the team's focus on that item, enhances the team's productivity, and increases the overall product quality.
It is understandable that when someone else does estimates for you, then you won't get that motivated compared to the dedication and focus you can achieve in case the estimation is done by yourself. It is because you will do all in your power to achieve the estimated targets knowing that it is you who have set that estimated targets beforehand.
When the whole team participates, it gives a gesture of teamwork and a collaborative environment that results in a quality product. Not all members are required to provide estimates, but the insightful question contribution from them can make a huge impact.
Now that we know how valuable it is to have the whole team take part in the estimation session, let's discuss a common issue or flaw in the Planning Poker technique.
As per the Planning Poker technique, participants are required to join the session at the same time, which means they have to delay whatever work they are doing currently. Once the session begins, they have to remain focused throughout the session because they first have to understand the product backlog items properly to provide the right estimates. So, members who are slow-learner might find it a bit challenging to understand the items on-spot and provide estimates. Similarly, members that already have a busy, tedious day might also struggle to focus properly in the estimation session. All these factors influence even worst for those members who have irrelevant skills because they get a feel of forced work, so their contribution part can impact greatly.
Since globalization and the global economy concept is becoming a new trend, it's normal to see team members working remotely or from different time zone. So, the concept of whole team participation in Planning Poker estimation session becomes a challenge for teams having remote members.
All the above challenges do not imply that you should skip the concept of whole team participation. Instead, you should find a solution. In this case, the best solution is to opt for Async Poker or Asynchronous Planning Poker. Async Poker does not just ensure that teams follow the Planning Poker technique, but it also removes the one-sit-estimation practice and remote working members issues.
Asynchronous Planning Poker or Async Poker can be termed as an advanced Planning Poker technique that is meant to address the concerns of distributed or remote agile teams. The working principle of Asynchronous Planning Poker is as follow:
- The product owner or moderator sends the details of product backlog items to be estimated to all the team members via email. In addition, the team members are provided with a deadline until when they have to submit their estimates or questions.
- Each member can study the items at his/her own pace and at whatever time suits him/her.
- Once the deadline reaches, all members send their estimates or questions to the moderator.
- The moderator then compiles the estimates. If the estimates match, then the team is informed about the final team's estimates. However, if there are significant mismatches or some members have raised important questions, then the moderator can set up a quick session (online session) to discuss the issues and reach a final consensus. Since all members already understand the items, so the session is usually brief, focused, and productive.
This way, distributed agile teams can participate as a whole using the Asynchronous Planning Poker technique. The key benefit of this approach is that everyone gets the time to understand the product backlog items at their own pace, while no one is forced to join and spend significant time to do the one-sit estimation.
This article tried to clear the misconception around whole team participation in a Planning Poker estimation session. It's true that irrelevant skilled members cannot bring much to the table, but their contribution during the discussion can be highly valuable. It is recommended to test out whole team participation for a couple of estimation sessions to see how it turns out for your team's productivity. And if your team is working remotely or from a different time zone, then do consider practicing the Async Poker technique.