The first thing to remember when it comes to estimating work is that it isn’t on a scrum master to come up with a number, a recommendation, or an estimate about anything on the backlog.
The development team are the experts and it is they who will discuss each backlog item, estimate how much work is involved in completing the task, and collaborate with others to decide which of the backlog items are ready for the sprint backlog.
That said, the scrum master can play an important role in the backlog refinement and sprint planning events by helping to facilitate discussions and conversations that empower the team to make great decisions around what needs doing, why it needs doing, and how best to get the job done.
Estimation is the practice of deciding, as a group, how much work or points is involved in a specific backlog item.
It’s an interesting event because people have different levels of experience and expertise when it comes to creating products or features.
For some people, a backlog item may seem relatively straightforward based on their previous experience and they would assign 3 points to the item. For others, the testing element may be far more complex than the item itself and for this reason, they would assign 5 points to the item.
As a scrum master, you want your team to individually consider each of the backlog items and do their own sums. You want them to think about the item, what all it entails, and how best to complete the item.
It is important that each voice is heard and respected.
Often, you will find that a majority of people will concur that the item is worth 5 points but you have an outlier that estimates 3 points whilst another estimates 13 points for the same item.
Discussion and conversation around these variances is incredibly important.
You want to ask the outliers why they have estimated the number they have and what their line of reasoning is behind the estimate they have given.
This kind of discussion opens alternatives and explores new thinking.
In some cases, the higher number has been estimated because someone has experience with the item and is able to articulate effectively why their estimate is correct based on past experience.
If the team agree on this, you are rewarded with an accurate estimation that ensures the team are able to achieve their sprint goals and objectives rather than underestimating the amount of work required.
You may also find that the person who estimated the lower number is able to complete the task extra quickly because they have vast and significant experience with items of that nature.
Again, this is great because the group are quickly and easily able to make a decision that the person with the most experience with that item should be assigned the task and the team will be able to ensure that the item is delivered quickly and effectively.
Planning poker is a great way to secure everyone’s estimates without anyone being swayed by an expert opinion.
Each member is given a deck of cards with numbers assigned to each card and when asked to present their estimate, the team all showcase their estimate in the form of the poker card number.
Why is this effective?
Because sometimes an experienced or senior developer may sway opinion beforehand by explaining their estimate. People in a group may have different ideas or have different experiences but will defer to the more senior individual because of team dynamics.
In planning poker, people are able to reflect on their own estimate based on their own level of expertise and experience, and safely share their estimate with the team.
As a scrum master, you can facilitate discussions around the estimates and have conversations around any differences that may crop up.
Through this process, the team collectively learn.
They get an insight into each other’s line of reasoning and explore the thinking behind how the backlog item can best be delivered.
It is inevitable that conflict of opinions will arise from time to time.
Conflict can be incredibly healthy and productive for a team if handled correctly with respect and openness.
Each of the ‘opponents’ are able to articulate their thinking, their line of reasoning, and what their preferred recommendation would be.
After that, you want to open that discussion to others and ask the two protagonists to avoid interfering in the group discussion.
Their opinion and recommendations have already been heard, now you want to give each member of the team an opportunity to voice their own line of reasoning and feel safe to make their own recommendations.
In many cases, a great line of reasoning will win the day and the team will agree on the matter without fuss or drama. That isn’t always the case though.
As a scrum master, you want to develop a way of making decisions when agreement isn’t unanimous.
Sometimes, people are going to hold strong beliefs and you simply won’t be able to find a middle ground.
In these circumstances, it is best if the team have a predetermined way to overcome these kinds of obstacles.
A common methodology is to hold an impromptu vote by showing hands with the majority of votes winning the day.
It is important that everyone is onboard with the decisions that are made regardless of whether it was their initial idea or whether they were initially opposed to the idea.
Having a predetermined methodology for dealing with these deadlocks ensures that the team move forward with respect and a sense of fairness.
As a scrum master, you play a crucial role in helping the team make great decisions around work that needs doing and how long they estimate the work will take to complete.
Follow the guidelines discussed and you will have a rock-solid starting point from which you can build.
If you like the idea of mentored and coach-driven skills development, visit our Agile Coach Academy page.
If you have identified coaching as a valuable skill to develop, visit our on-demand Introduction to coaching course page.
For more information on John McFadyen, visit https://www.growingscrummasters.com