Tracking Time at Work: Planned vs. Unplanned Work
We've all been there. You plop down at your desk, bright-eyed and bushy-tailed 🐿️, coffee steam curling lazily up from your "best dog mom" mug ☕. You know what you need to do today.
You loosen your neck, flex your fingers, open up that first application on your desktop and....
Ping....
Someone has pinged you. They need this other thing done asap. Ok, no problem, you can pivot, you're flexible. Ten minutes into refining exactly what this person needs from you on this new project, you hear another....
Ping....
And now you know exactly what type of day this is going to be.
Planned vs. Unplanned work
After having a measure on how much work I do, the next level of refinement (in my mind) is how much of that work is planned work vs. unplanned work.
Definition
All work is either planned or unplanned. For this project, I defined unplanned work as any task I worked on that I had not anticipated needing to do at the start of the day.
To help keep track of this, at the start of each day I planned out which tasks I needed to complete that day and which projects I planned to make progress on.
Anything else that cropped up during the day that did not fit within this starting list of activities fell into the category of unplanned work.
Why measure this?
For myself I have observed that unplanned work has the greatest negative influence on my mental fatigue and on my ability to accurately estimate how long it will take to perform a task.
Unplanned work typically correlates strongly with interruptions, interruptions create focus churn, and focus churn yields mental fatigue.
When it comes to estimating the time needed to complete a task, it is helpful to know how much padding you should add to that estimate to account for unplanned work that may come up. What may take you four hours of uninterrupted time to complete, could easily turn into two days if you aren't able to get more than 20min here and there to work on it.
The Data
The following data was measured over 59 working days.
Roughly 20% of my time is spent on unplanned work.
Let's see if we can identify where this unplanned work is coming from.
Does Location matter?
Currently there are two locations where I do work, my home and the office. Naturally I was curious to see if being physically around others, where it's much easier for people to just drop by your desk, impacts the amount of unplanned work that happens.
| Location | % Unplanned work |
|---|---|
| Home | 19% |
| Office | 27% |
On days I'm in the office there is a 42% increase in the amount of unplanned work that crops up.
Who is generating all of this unplanned work?
The main stakeholders at play here can be described as:
- My Team / My Teammate - People directly on the same team I belong to.
- Team [X] - Some other team
- My Pod - A pod is a collection of engineers across many teams who can represent a full-stack development team focused on end-to-end feature delivery. I lead one of these pods.
- Pod [X] - Another Pod.
- Product Manager - Each Pod has a PM who is setting the priorities and roadmap for that pod.
- Everyone Else - The long tail of stakeholders each generating less than 4% of the unplanned work
In general, most unplanned work is done for the benefit of my Pod or my Team. While I would like to minimize unplanned work, at least the majority of it seems to be for the benefit of those within my direct areas of responsibility.
I also thought it would be interesting to compare how some of these numbers change by location:
| Stakeholder | Home | Office | Diff | % of Stakeholder who is Full Time Remote |
|---|---|---|---|---|
| My Team | 19% | 2% | -17 | 50% |
| Pod A | 7% | 0 | -7 | 100% |
| My Teammate C | 4% | 3% | -1 | 0% |
| My Direct Manager | 7% | 5% | -2 | 0% |
| My Teammate A | 7% | 8% | +1 | 0% |
| My Teammate B | 6% | 7% | +1 | 100% |
| My Product Manager | 6% | 7% | +1 | 0% |
| Team C | 6% | 5% | +1 | 50% |
| Myself | 2% | 7% | +5 | 0% |
| My Pod | 10% | 16% | +6 | 50% |
This data was surprising to me and therefore also very interesting. I added the column % of Stakeholder who is Full Time Remote to help me see if any pattern emerged.
Not everyone at my company is required to come into the office, so even on in-office days, I spend the majority of my time on zoom calls or on slack interfacing with full time remote employees.
I was hoping for the above chart to reveal that when in the office, I had more unplanned work from people who were also in the office with me. But that does not seem to be the case.
I think its safe to say that at this time Stakeholder x Location is not a useful intersection.
What kinds of projects generated the most unplanned work?
Once again the `Everything Else` bucket is a long tail of small slivers each representing less than 3% of the chart. Excluding this catch all bucket, the majority of my unplanned work seems to come from people asking me questions. After that, we see a collection of various projects that generated unplanned work.
Let's talk about Project A for a second...
As we can see, Project A was one of the largest contributors of unplanned work during this time frame. Project A did not go well.
This was probably one of the roughest projects of the year, things went wrong with it constantly, and I had many, many learnings from it.
43% of work done on Project A was unplanned. Let's compare that to a few other projects I've worked on recently. I'll also add in my perception of how smoothly these projects went.
| Project | % Unplanned Work | Perception |
|---|---|---|
| Project A | 43% | Oxygen masks have been deployed |
| Project F | 27% | Keep your seatbelt buckled |
| Project G | 7% | Feel free to move about the cabin |
Does more unplanned work happen on certain days?
Wednesday is definitely looking like a hump day to me 🐫.
The week seems to get off to a good start with lots of planned work getting done, but by Tuesday there's already a sharp uptick in unplanned work, and that increases even further by Wednesday.
Thursday's seem to rally and by Friday we're very close to our starting point at the beginning of the week but haven't fully recovered.
My first impression upon seeing this data is that we as a company cannot stick to and execute on a plan. That what we set out to accomplish at the beginning of the week ends up not being highest priority. We start pivoting Tuesday through Wednesday, and have realigned by the end of the week onto new priorities.
However, another way to interpret this data would be that as people start working on their planned projects at the beginning of the week they run into blockers and questions, requiring them to loop in others who had not been planning to contribute. This culminates around the middle of the week, and as people make further progress on their projects they have fewer questions and blockers.
There are several other hypothesis that come to mind, and I suspect they all play a role in this problem without any single one being the main culprit. But this trend is interesting and definitely worth investigating more.
Does more unplanned work happen with a certain AOR?
At first, I was shocked that my role as an Architect would be where most of my unplanned work originates, but as I thought on it more, this makes sense. While I have plenty of planned projects to do within my capacity as an Architect, I am also responsible for hunting down weird bugs, investigating system issues in Production, and chiming in on team technical discussions. Most of the things in this list are ad hoc, unexpected, and unplanned for.
Conclusion
There is quite a bit of data to sift through when it comes to Planned vs. Unplanned work, on top of that there are many ways to slice and cross-compare this data. I shared here the bits that I either found most interesting or most actionable.
What levers can be pulled on?
- Significantly more unplanned work pops up when in the office. If I need to make uninterrupted progress on a single project, avoiding the office will decrease the chances of unplanned work popping up.
- Risky projects yield more unplanned work. It is important to invest enough planning and preparation up front on a project to ensure it runs smoothly. Additionally, be on the look out for a project starts going south.
- Unplanned work seems to follow a pattern over the week. I will make the most progress on planned projects if I work on them on Monday's and Fridays.
Areas for additional investigation
- Attempt to identify what causes the "hump" pattern where the most unplanned work happens towards the middle of the week.
- Dig deeper into sources of unplanned work in my role as an Architect. If the bulk is related to investigating broken things, then that would be good to know.
- A large portion of unplanned work comes from queries. Dig deeper into this and see if any patterns emerge amongst these questions.