posted 7th February 2024
Intro
With a lot of the companies I work with, their systems are complex, where work spans multiple teams to deliver real customer value. Everyone seems to agree that we want flow so we can meet our customer needs early and often. But are we helping ourselves or hurting ourselves in the wider organisation in many of these cases?
In many of these companies, we may be limiting Work in Progress (WiP) at the Portfolio level. We may even be limiting WiP at the leam level and starting to see lead times reduce (whoooohoooo!). Even at the team of teams level (not always as often I'll add). Yet we don't limit the WiP and manage the flow at the Epic level. Why not?
Let's dive in and explore this!
So what is an Epic?
An Epic is typically a large body of work that can be broken down into a number of smaller units of work. Epics often encompass multiple teams, possibly on multiple projects. Epics are almost always delivered over a number of cadences. So these are effectively bigger pieces of work and will often deliver large pieces of value for the business.
The problem and how it hurts us
If we don't manage the system at this Epic level it is easy for us to have lots of Epics open and when we do finally inspect our system it will most likely seem blocked. Or at least we no longer know what our priority is. If everything is important, then nothing can be right.
Also, we don't always realise meaningful value, especially if we have component teams or our teams are not fully cross-functional. Our Senior leaders may not see the value we thought these Epics would bring. If they don't close, how can they?
We talk about lead times reducing at the team level which is great. We'll probably celebrate these wins. We know about the portfolio and what we aimed to invest in. However is the work meaningful enough and ultimately aligned enough at each level for our customers and leaders to see a real difference? If we deliver 10% of Epic X, 15 % of Epic Y, and 10 of Epic Z, did we make a dent in our strategic goals? It's like having 4 lanes of a bridge partially complete but no one can cross the bridge so there is no value until we can get to the other side.
But we have Epics as containers and it works for us
What about epics as containers? That is how we have always used them. Tech Debt is a common one.
Yes, it can work but how do you know the impact of your work? Also, is it an Epic we align work to that we just kick the can down the road on? If there the epic is never closing then where is the hurry? What is the priority call? If my team is busy on Epic A and your team is busy on Epic B we can just plod on and deliver sprint after sprint. Maybe next sprint we can do a bit on Epic C and we all feel great about meeting our sprint goal.
However, what if there is £1 million of profit we are not realising in Epic A? Knowing that information would help us make the informed choice and focus our effort there to realise that £1M sooner. If we don't understand what is needed to get Epic A, B, or C over the line then it's unlikely we know what priority order they needed in as we never close them and the system just gets further bloated at this level. If we keep adding cars to the road we end up in a traffic jam where nothing is moving.
What can we do about it?
Whilst there are several answers here, these are three that jump straight to mind:
Visualisation: Do we make this level of work visible? If not, maybe we need to consider having Epic boards that we regularly review.
Someone to manage the flow: Maybe we need someone to manage this level of work for us? From experience, when we have someone actively managing the work at the Epic level we gain more focus and prioritisation is easier. Alignment at the team of team levels is clearer and empowerment to say no is better as everyone is now on the same page.
Acceptance Criteria: Do we collectively understand What needs to happen to close these epics? Maybe we need to agree on clearly understood acceptance criteria before we open these Epics. If we have this then making an informed decision on when to start these, when to close these, and shift focus becomes way easier in the future.
What happens when we start to manage the work at this level?
I see this as a common theme and once the focus comes to this level of work, we then start to ramp up on so many points. Noticeable and tangible flow really begins. Alignment grows along with a sense of achievement across the departments and across multiple teams. Impact is more visible and other decisions are easier both at the portfolio level and at the team level. Collaboration across teams will increase. Yes, difficult conversations will most likely be needed more but we were probably avoiding these in the past because of a multitude of reasons. We did not need them as we were all doing our own thing with our team that yes aligned to an Epic but were we the only team on that epic when multiple teams are needed to meet the currently unknown need as we did not have a clear understanding of what was needed.
Summary
Applying some of the general practices of the Kanban Method can help you here. Visualise the works so that we know what is in our system at all levels. Actively manage flow so we know we are delivering what we intended to and we can take action if not. Applying acceptance criteria to our Epics so that we know when to close them and we have met the need. These are just three of the six general practices and they can help you so much. If you want to learn more on this, please get in touch.
What has your experience of working with Epics been like? Do you actively manage them and was it always this way? Were things better or worse when you did not manage them? I'd love to hear your thoughts on this.