Workloads #2 — How hard does Adam really work?

“What’s worse is that long hours, excessive busyness, and lack of sleep have become a badge of honour for many people these days. Sustained exhaustion is not a badge of honour, it’s a mark of stupidity.“ — Jason Fried, It doesn’t have to be crazy at work.

As academics we all well and truly know the aspect of workplace culture being described in the quote above. I’ve been there myself, working days, nights, weekends, to the exclusion of most other things to ‘advance my career’, and then trying to use that as a badge of honour by humblebrag, as the vast majority of academics do. At first all these hours seem fine, but eventually they start to wear you down, and ultimately destroy you.

Inspiration — How this all started: When I went on sabbatical in the 2nd half of 2019, I committed to not working at all on weekends, limiting work in the evenings and not being in the lab beyond 8am-6pm — something of a challenge in Japan — in the hopes of staving off exhaustion. This helped a little, but when COVID-19 took off in 2020, the hours became insane again — at its worst I was up 7am, clocking off around midnight, and sometimes working on the weekends as well, just to hold it all together, between managing the school’s teaching, rapidly shifting my courses online, and trying to keep my research group alive.

By the time 2022 came around, I had well and truly had it. There was no longer any denying that I was deep into burnout, as many of my colleagues were, and struggling to hold productivity constant, let alone at the endlessly rising levels being expected from above. I was exhausted, depressed and seriously unhappy. Change was needed, dire change, or I was going to end up in a wooden box having let my whole life drift past while being too busy doing things that weren’t even seeming to be appreciated, or even noticed, let alone being paid for (e.g. editorial & refereeing tasks).

“If you try to find time for your most valued activities by first dealing with all the other important demands on your time, in the hope that there’ll be some left over at the end, you’ll be disappointed.“ — Oliver Burkeman, Four thousand weeks: Time management for mortals.

I spent my scant spare time during covid lock-down (an academic have spare time? how dare they!) reading books like Oliver Burkeman’s Four Thousand Weeks and Jason Fried & David Heinemeier Hansson’s It doesn’t have to be crazy at work, and it inspired me strongly to fight for solutions. From April 2022, I started to track my working hours by project and work-mode down to five-minute blocks. I wanted to understand where my time was going so that I could optimise for effectiveness & efficiently more ruthlessly than ever before. Not necessarily to do more than ever before and thereby win the academic metrics arms race, but instead to get all I could out of a sensible day and then get back to having a life. I know this operating mode is ultimately not sustainable from an academic career perspective — how I’m dealing with that is a topic for an upcoming blog-post.

Before getting into the detail of this, I need to acknowledge my former colleague Christine Lindstrom for building ‘Version One’ of this spreadsheet back in late 2021. Seeing the potential in it, I built a variant of it for my own use and it has continued evolving since I started tracking my own time in April 2022. I’ll provide my version with this post so you can do the same.

Now that I’ve been doing this for well over a year, I’d never go back to not tracking my work time in detail like this — it is by far the best way to be conscious of how I spend (or waste) my work hours, and inform the decisions of what to do and what not to do to make sure my work hours don’t get much beyond what I’m paid for unless there’s a very clear benefit to me personally in doing so.

“Productivity is a trap. Becoming more efficient just makes you more rushed, and trying to clear the decks simply makes them fill up again faster. Nobody in the history of humanity has ever achieved ‘work-life balance’, whatever that might be, and you certainly won’t get there by copying the ‘six things successful people do before 7:00am’. The day will never arrive when you finally have everything under control — when the flood of emails has been contained; when your to-do lists have stopped getting longer; when you’re meeting all your obligations at work and in your home life; when nobody’s angry with you for missing a deadline or dropping the ball; and when the fully optimised person you’ve become can turn, at long last, to the things life is really supposed to be about. Let’s start by admitting defeat: none of this is ever going to happen. But, you know what? That’s excellent news.“ — Oliver Burkeman, Four thousand weeks: Time management for mortals.

Down to brass tacks: Immediately following are two links to my current hours sheet (excel files) — the first has one month of data in it (April 2023) so you can see how it actually works; the second is completely empty and ready for use by anyone who wants to try it for themselves. I offer this under CC-BY terms and welcome both its use and evolution across the higher education sector. The data provided for April 2023 is real data; the only thing that has been modified there is to slightly de-identify who I had meetings with. The same goes for all the figures I show below as well.

First thing you’ll notice, nothing shown here has more than 35-40 hours a week, and that’s not because I’m hiding it to avoid glorifying over-work. I really am that disciplined now, but it’s a double-edged sword — at the same time I must be utterly single-minded (almost to the point of ruthlessness) about using every minute of those hours wisely. Some people laugh in disbelief when I tell them that I only work 35-40 hours a week now, but I do, it’s really that simple. On average I’m only 5% over the 35 hour week that I get paid for, and looking across my year and a half of data, the peak was 21% over (42.5hr in total).

“There is nothing so useless as doing efficiently that which should not be done at all.“ — Jason Fried, It doesn’t have to be crazy at work.

The details: What does my system look like? The main part of the architecture, in terms of daily interaction, is the monthly sheets. These are where the data is collected, by days and weeks, with a typical week shown below.

Tasks are listed by start and end time, with a column that autocalculates the duration, and then tracking by work-mode (e.g., O for office work, P for presentation, M for meeting — full list given above) and project (e.g. T-2113 for teaching in PHYS2113 Classical Mechanics, R-Bio for my ARC Discovery Bioelectronics project, R-PER for my physics education research work, etc. — more on projects later), along with a short description of the task. Note that in some things, especially meetings, I’m actually multitasking, but since the meeting has priority, it’s in the timesheet — all it generally does is cut down email allocation (Project O-e) since that usually what the multitask is. The system automatically tallies up the hours for the week and feeds this to other parts of the sheet. The first part is the top of each monthly page, which has a nice distillation of hours divided by project and work-mode, as shown below.

You can see that I spent most of June doing teaching (T-2113 and T-SR) and research (largely hands-on for R-Bio) with some admin (mostly email in O-e) and some grant reviewing (S-ARCRef) added in. A nice feature of the pie charts is that you can see very clearly where your time is going, and it provides a big incentive to maximise the things that matter and minimise the things that don’t as the months progress.

Projects: Before getting to annual tallying I need to deal with projects, which is the first of the 14 sheets in the excel file. I’ll come back to the top of the ‘Projects’ sheet below, but scrolling down shows details of all the projects I can allocate work to. These are broken into four sub-sets: Research (R), Teaching (T), Service (S) and Other (O). I’ll give two examples here, but you can see them all in the attached files.

The first example is research, where I split my various projects: the first is a lead-CI ARC Discovery and a NZ Marsden I’m AI on (with Natalie Plank and Colm Carraher) and the second is an ARC Discovery I’m second CI on with Lawrence Lee. These have huge hours allocated in planning because they are major priorities, and if I can put more than 100% of my allocated time against those, it’s a win (providing I’m not compromising my teaching quality or overworking in the process, mostly it’s clawed back admin time by being militant about meeting avoidance). The rest are either projects that have ended (e.g., R-Naf or R-MoF) and need final reporting/closure, or wider collaborations that I contribute some small time to (e.g., R-PER or R-Side).

The second example is Other, which contains everything from email time — often hard & absurd to pin to a specific project, hence my daily O-e ‘General Clear’ task — to OH&S tasks, meetings and departmental social functions (they’re work too, right? If it wasn’t work, I wouldn’t be there). There’s an allocation for managing the timesheet itself, mostly annual turnaround and maintenance. The day-to-day running amounts to <5 min/day. There’s also (amusingly) a category for pointless bureaucracy (pronounced /ˈbʊlʃɪt/) for when, say, faculty makes me do pointless data entry tasks that others should be doing, surveys, etc. In contrast to research, the ‘pointless bureaucracy’ project is one that I do observe a very hard 100% limit on — once I hit that limit (for 2023 already well before September) not only do I not do any further pointless bureaucracy, I don’t even reply to requests to do it.

The answer is not more hours, it’s less bullshit.“ — David Heinemeier Hansson, It doesn’t have to be crazy at work.

Year planning: The top of the Projects sheet is where all the annual planning happens. It looks like this:

Every project, grouped by block, appears here, and the columns are the planned hour allocation at the start of the year, the actual hours accumulated, the proportion of my annual hours (1610 hours) that this represents, the current actual-to-planned hours ratio, and the rate I’m accumulating hours relative to pro-rata for the year (<100% than actual will be under planned at year end, >100% than actual will be more than planned at year end). Thereafter, I just have the feeds from the monthly sheets, so I know the time per project per month (good for deeper insights).

Before considering this in more detail, it’s important to show where the planned block totals (Research 523 hours, Teaching 518 hours, Service 255 hours and Other 314 hours) come from. This calculation is done on the ‘General Rules’ sheet as follows:

I basically count the contracted hours for the year, divide this 40:40:20 between Research, Teaching and Service, and then tax each category 20% for allocations to Other. Then, within each block I allocate hours to the various projects as I see fit/sensible. The one exception to this rule is Teaching, where I am guided by the task splits from the school’s WLU scheme as to how I allocate my time. You’ll remember in Workloads #1 that I said a difficulty was that WLU-to-hours conversion was weakly defined — here I fix that, because for everything to work, I need it to be 1 WLU = 1.10 hours, and that ultimately sets how I plan my teaching time to make everything manageable. This more or less makes the WLU conversion a good ‘price signal’ that quantitatively governs my resource allocation (i.e., time) to the various tasks I’m given — this is why it’s very important for that number to be sensibly, accurately and globally (at school level) estimated. If the WLU conversion is absurd, then any attempt to sensibly manage academic workloads in a school becomes nothing but a joke (or negligent management in my view).

Ultimately, you’ll notice that the actual numbers I devote to tasks as the year progresses can deviate from the planned numbers, and that’s something I actively decide on the fly as the year progresses. In running a system like this, one has to walk a line between having no system at all and having a system that’s so strict that you compromise your ability to optimally function. That line is ultimately a quantitative system like this with some sensible tolerance for ‘noise’ and adaptability as the year progresses. Don’t let perfect be the enemy of the good.

“If the only way you can inspire the troops is by a regimen of exhaustion, it’s time to look for some deeper substance. Because what trickles down is less likely to be admiration but fear and dread instead. A leader who sets an example of self-sacrifice can’t help but ask self-sacrifice of others.“ — Jason Fried, It doesn’t have to be crazy at work.

Year Overview: The final tab in the sheet is the Year Overview tab and this appears as follows.

The tab has each week of the year allocated to a row. From left to right, the columns are the ‘Monday date’ (if only we had the Swedish week counting system in Australia), the calendarized full-time hours for the week (i.e., corrected for public holidays), the scheduled full-time hours for the week (i.e., corrected for any leave I take), and the actual hours I work in that week, which is taken automatically from the monthly sheets. In other words, this tab automatically updates as the year progresses — the only time overhead for it is the initial set-up at the start of year (i.e., adjusting to the next year’s calendar).

There are three calculated columns that follow this. The first is the relative load column, which tells me how my actual hours worked compare to my scheduled hours. This column is colour-coded such that it is pure green below 100%, but shades of red creep in beyond 100%. The second column is the over/under column, which is just the actual hours worked minus the scheduled hours, it’s primarily there to feed the column to the right of it. The third column is the balance of hours under or over-worked and is essentially a ledger of how many hours I owe UNSW (negative balance) or UNSW owes me (positive balance). The starting balance is 134.7 hours because I roll this continuously from year to year, having started at zero when I started this hour tracking system (much of this was accumulated while I had a large admin role and before this system made me ruthless about eliminating academic ‘make work’ from my work activities). Had I started this system when I started in 2003, the balance would likely be approaching ten thousand by now. Even if I’m as massively disciplined about my hours as I am this year, the annual balance grows at approximately 100 hours/year. I currently consider this balance to be something like ‘flexi-time’, in the sense that I can work harder one week to claw back some free time the following week for appointments or an early-finish, even perhaps take the odd extra day of leave using the overworked hours, which I sometimes (but rarely) do.

“Time-management hacks, life hacks, sleep hacks, work hacks. These all reflect an obsession with trying to squeeze more time out of the day, but rearranging your daily patterns to find more time for work isn’t the problem. Too much shit to do is the problem.“ — Jason Fried, It doesn’t have to be crazy at work.

Closing thoughts: At this point I can envision a huge spectrum of reception to this post ranging from the ‘How dare he! This is unprofessional for an academic!‘ (esp. at the ‘flexitime’ idea I just mentioned) to ‘This is incredible, I must start doing this too.‘

Let me try to condense the former across to the latter by stating that this has been, by far, the best productivity move that I have made in my entire 20 year academic career, and trust me, I’ve tried a lot of productivity hacks over those 20 years.

The spreadsheet simultaneously enables a laser-like focus on what truly matters to generating real productivity (and not the imaginary productivity of academia, which I’ve had to ruthlessly trim to make my hours add up) and provides me with sufficient down-time and rest that when I am at work I am 110% on task and ‘firing on all cylinders’. My ability to come in, go hard, and then go home is better than it has ever been, and at the cost of a few minutes a day for the accounting that puts a quantitative backbone to the internal ‘call for discipline’ that ensures my effectiveness and efficiency.

Some might also say “This is nice, but what stops you from ‘cooking’ your hours? How can we trust you?” It’s a good question and there’s ultimately two answers to it. The first is that the only person who sees this sheet is me — if I cook the numbers all I’m doing is lying to myself. Why would I lie to myself? What’s the point of that? The second is that the true benefits of the sheet come only if I’m completely honest with the sheet — I’d just compromise the benefits by cooking the numbers. Indeed, this is one of the motivators for the ‘flexitime’ aspect of the sheet, for me. If I overcook my hours, I remove time I can spend on actual productive work, and if you are in your job for the right reasons, then you really do want to do actual productive work, right? And if I undercook my hours, then I pitch myself back towards burnout and exhaustion and all the unhappiness that it entails.

So, I encourage people to try this model for themselves if they want. I’d be happy to receive feedback from people once they’ve tried it for a while. Feel free to edit, adapt, tweak, etc. It works optimally for me, but I adjusted it from Christine’s version to achieve that, and I can easily imagine others would need to adjust from my version to make it work for them also.

In the third & last post on workloads, I’ll look at workloads above the school level for some ideas on what we could be doing there to improve workload management by being more quantitative at the other levels.