ProdGoal Logo

Why Forecast? Understanding Everyone’s Role In Kanban Forecasting

Why Should We Even Forecast?

There is so much fuss around forecasting that it’s quite reasonable to ask why do we even do it? Why should we forecast? Why can’t we save everyone’s time and just say that “It will be done when it’s done”? Especially since this is the only true forecasting statement there is.

Why Forecast?

While this is a very important view that we need to keep in our mind, there is still a strong case for making forecasts. Their value lies in having an idea of when something will be done and the value is at least twofold.

First of all, the duration is giving you information about the cost of the work. And the date is giving you invaluable information about how to synchronize with each other. Without this, it’s very difficult to run a viable business or to synchronize multiple teams, deparmants or just parts of the organization.

One of the clearest forecasting examples is a question from a salesperson who wants to know when something will be done so that they can sell it then or try to keep the prospect in the loop until that date. If they don’t know the date or if the date is way off, they can lose this deal as well as their trust.

If you are not sure about this, just imagine committing to buying things without knowing their price, which would represent the duration of the work. What would that feel and look like? Or imagine that you need to meet with multiple people for lunch, but you don’t know when they will be able to make it. Doesn’t sound like a proposal that you would accept. You need to know when they will be able to make it so that you can organize yourself as well.

On the other hand, it’s understandable why people don’t want to forecast. There are so many bad experiences out there on all sides… The biggest clash points are around the accuracy of the forecast.

Everyone knows that new work can be added or discovered. Some crucial parts of work can get postponed or some hard dependency can put everything to a stop. There are always things that you can change, but will this keep you on track? And this can happen even if you are the most Lean and Agile team there is.

But this still doesn’t mean that we shouldn’t forecast and just go with “It will be done when it’s done”. This is probably the only forecast that will be 100% true, but it will still be 100% useless since it can’t help us evaluate the cost, synchronize people or teams or make good decisions.

To sum it up, if we don’t forecast, we won’t “lose” since we are not giving the right forecast. But this way, nobody will win either. They will not be informed in any way and this won’t do anyone any good. The only way to truly succeed is to understand each other’s roles, embrace them and communicate and collaborate on our way to success.

Kanban Forecasting And Communication

When doing Kanban forecasting, people usually think that creating the right forecast is the biggest problem. This is partially right because there are many factors out there that you need to take into account to do a good forecast. There are different techniques and models to do this, but even when you create a forecast that’s only the beginning.

Most of us work in dynamic environments with a lot of moving parts. Things change. Scope changes date change, and team members change. Everything changes and so does the forecast. We have put so much effort into creating it and tried to take all factors into account, and yet, it seems like it’s going to change. What do we do now?

We need to communicate properly. And by communicating, I don’t mean that we just need to find a nice and polite way to say that we are going to be late. I mean that we need to talk about the right things, ask the right questions, and make the right decisions to make the forecast successful. We truly need to master the forecasting communication.

When Will It Be Done?

As you can see, communication in Kanban forecasting does not mean that you only communicate the end date, but you will certainly get asked that.

Whether you are using Kanban, Scrum, or some other Agile framework, people will want to know at least one thing about the task you are working on. And that thing is the dreaded question - When will it be done?

Let me tell you that this is not only a dreaded question for the person being asked. This is a dreaded question for everybody. Why? Because everyone has a stake in the answer that will be given. But even more so because people think that they can define the date and then go away and things will happily end on their own.

Nope. This is at least not how most cases end. We can give better or worse forecasts using different forecasting techniques and communication patterns. On the other hand, it’s still very difficult to give a correct date that will be defined long in the past and that will happen exactly when it was planned.

We all need to let go of this illusion. What we need to do is to communicate and collaborate towards the desired date and make it happen. This is the only way that will truly get us to do our best to deliver on time. Everything else is some old forecasting fairy tales that we still believe in.

I will repeat and emphasize that we all have a stake in the forecast. We can’t just delegate it by asking the magic question “When will it be done?”. Getting the answer doesn’t mean getting to success.

”When Will It Be Done?” And People Committing To A Date

Some people think this question will result in a date and thus a commitment that will make things happen. I do understand the part that we want people to feel committed and responsible. This kind of question can help with that. But it still won’t make things happen.

When it comes to forecasting, it’s not only about the people giving the forecast. It’s about the whole system and all people that are included in the system. If the system is not being managed and if all parts are not communicating and working together towards the same goal, one commitment doesn’t mean much. Every nonoptimal system can eat any forecast for breakfast.

If we want to create a good system and have the right communication and collaboration flow, the first step is to understand the various sides that exist in a system. When each side understands the other side, they start to see the shared goal. And this is one of the strongest motivators for communication and collaboration.

Forecasting From The Eyes Of The Person Giving The Forecast

If you are the person doing the work, you know that you are not sure when exactly something will be done. You know that some unexpected additional work can come out of it.

You also know that there is a chance of some very urgent work suddenly appearing out of the blue. If that were to happen, everyone will expect you to finish this new work ASAP, but it will mess up the dates you’ve communicated and you’ll be late.

There is also a chance that a crucial person will get sick or that some other dependency that you can’t affect will get you off track. And you still need to give an estimate, maybe even right now.

You don’t want to be too pessimistic in your forecast because it can look like you are slacking. Or that you are not proficient in your job or that you are just giving yourself too much time because you are afraid to fail.

On the other hand, it’s also dangerous to be optimistic. You have a request and as a good worker, you want to say yes to it and please people.

If you know the date when they are expecting it, it’s even worse. Now you don’t want to disappoint them and seem incapable and unprofessional. You can just say yes to the date you know they want to hear and hope for the best. You know that if you fail and try to explain yourself, nobody will care. You will just look bad.

Or is this the case? Let’s also take a look at the other person asking for a forecast.

Forecasting From The Eyes Of The Person Asking For A Forecast

If you are the person asking for a forecast, you already know what trouble that brings and you can already feel the headache. You are also aware that no forecast will be precise, and even if it’s precise it probably won’t be accurate (or at least I hope you’ve figured this out by now).

You need to work with dates that other people give you, but you are not sure how the forecast is being made nor how much you can count on it.

There is probably already a date in your head - when you think the work can be done. But this is also just your forecast. Or you have a date when all work needs to be done, but it’s a date you’ve been given. A deadline. And it’s not the prettiest word out there. It comes from prison terminology. It means a line that if you cross - you will get shot. Very motivating.

Your job is to make things happen and to make sure that everything is on time, but you are as dependent as everyone giving the forecast. You’ve tried multiple ways to forecast before, but that didn’t go too well. You would like to have a more reliable way to forecast.

When these two are put together side by side, very similar struggles emerge.

And I say this, hoping that the person asking for a forecast is aware enough that plans change, new things arise, developers have limited capacity and scope will probably need to be cut to deliver by some target date.

To meet the dates we forecast, we need to be adaptable, communicate, and collaborate. Things happen.

Are We Worrying Too Much?

We will talk about probability quite a lot in the story of Kanban forecasting. And I will start by telling you immediately that if you recognize the worries mentioned here, there is over 90% probability that you see the situation much worse than it really is.

In my experience, there are very few people who don’t understand that you need to adapt to succeed. Most people understand reality very well and want to work towards a shared solution. What they struggle with is to find the right communication format and forecasting technique to do this.

If you use the right Kanban forecasting methods, there is a great chance that you will succeed and get rid of all those headaches. And go back to Kanban probabilistic forecasting, there is a 100% probability that there are better ways to forecast than those you are used to. And not only that – there are far better ways to communicate about forecasts and not have stress while doing it.

Since this topic has been deeply covered in another article, I would suggest that you check it out. There are various forecasting models and techniques out there and you should be able to decide which one fits best your needs.

A Word Of Caution To The Role Of Management

In my work, I have come to find that people making the forecast can change their ways of working and communicating and that’s not too big of a problem. The real problem is when this happens and there is no support from management.

Supporting, enabling, and escalating when needed are probably the most important roles of management in forecasting. If people can’t solve things on their own and are asking for help, you need to step in and do your part of the work or things will start to fall apart and people will start to leave.

Since you are the manager, you can also delegate this to someone. Or you can teach people how they should behave in this situation. Do whatever you need and what makes sense. But one thing you have to do is to embrace your role and act accordingly until the situation is resolved. Unless, of course, you want everyone to fail, which I’m sure you don’t.

”With great power comes great responsibility” is something that you should live by. Even if someone else has made a mistake, you have to own it and see what to do about it. If you are a manager, your responsibilities are higher but don’t worry too much, everyone needs to own their part of the forecast.

Extreme Ownership In Forecasting

One of the best concepts that I’ve encountered in my career is Extreme ownership. In essence, it means that everyone always needs to see what their responsibility is in each situation and own it. And they need to do it to the extreme.

If you want to be successful in forecasting, this is key. As we saw, each role has a part to play and there are so many situations where things can go wrong. We don’t need to point fingers at each other, but to own the situation and do whatever we can to succeed.

This is the mindset of the winners and extreme ownership needs to become a part of your company’s culture. Have in mind that it’s best if leadership and management become role models in this since it’s the most promising path to making the whole company embrace it. If leaders behave like this, everyone else will follow.

Whole System And Kanban Forecasting

As you can see, there are many things that everyone can do. So, no matter in which position you are or what role you have, you can do a lot. Keep in mind that Kanban forecasting is a collaborative sport and that it’s endlessly easier when everyone is on the same page and on the same side, trying to achieve the same goal.

In this way, Kanban forecasting is a team and company activity, whether you like it or not. And like in all group activities, it can be a virtuous circle or a vicious circle and everyone has a role to play.

Teams should be working on providing the best Kanban forecasting model, but also on communicating right, alerting on time, and making sure that they are doing their best considering their domain. It’s also very important that the team focus on doing discovery and delivery right and managing dependencies in a good way since these things can affect the forecast a lot.

On the management and leadership side there needs to be an understanding of the process, changes, and problems during the process. But also there should be a real support in solving problems that the teams can’t solve on their own. These could be questions about what to prioritize, what can be removed or delayed from the scope, or help with solving certain dependencies, both internal and external.

With all this in mind – understand your and each other’s roles, embrace extreme ownership and make sure you are using the best Kanban forecasting techniques. Some tools can help you with that and Prodgoal is a free online Kanban tool that has various advanced Kanban forecasting features. Feel free to check it out.