Book review: Overload by Kelly and Moen

Posted by Patrick Lam on Thursday, January 14, 2021

I was wandering through a Wellington City Library branch and picked up Overload: How Good Jobs Went Bad and What We Can Do about It by Erin L. Kelly and Phyllis Moen, a book published March 2020. I don’t usually read management books but I am interested in how people work in 2021; work-life balance was cited in Minister Navdeep Bains’s recent resignation, for instance (of course it always is for politicians), and discussed in a Globe and Mail Opinion.

Summary

This book describes the authors' experience introducing STAR, a program for re-designing work (“dual-agenda work redesign”) so that it better supports rather than supplants workers' lives. STAR was implemented as a pilot project in the IT workforce at a pseudonymous company TOMO, and ultimately revoked upon TOMO’s merger with also-pseudonymous ZZT.

  • The first part of the book talks about why modern work sucks for this workforce. This book is obviously a pre-COVID study and an important source of woes is complaints about required but not-useful physical presence, especially at meetings. Many workers suffered health problems which are plausibly a result of stress, lack of sleep, etc, consistent with descriptions from Millennial Revolution. I also didn’t know that people multiplexed virtual meetings even pre-COVID. The authors also discuss how this situation arose at this company; I provide commentary below.

  • Of course the authors have an agenda: the dual-agenda work redesign (STAR). I’d summarize it as empowering workers to decide how to best do their work. Sounds obvious, but many managers and coworkers think that workers only do work when they’re in the office. STAR cuts managers out of the loop on these decisions.

I’ve had conversations with friends about their PhD students not being in the office. I don’t believe in that, and I think that we’ll see lots of unproductive office time post-COVID (although it is definitely true that some people like being in the office some of the time).

  • The book then summarizes their evaluation of STAR’s effectiveness: basically huge increase in morale and probably retention; no effect on productivity. But finally the acquiring company was stuck in its old-school ways and didn’t buy it. (There are so many terrible stories in this pandemic January about workplaces in Ontario which are forcing workers to show up for no good reason.)

My Take on TOMO

I know lots of relevant context here. My students graduate and work in the IT industry. And, although I don’t do research in software engineering processes, I’m certainly familiar with that subfield. I only have first-hand experience working as an academic, but it’s also not like academics don’t have a lot of work.

TOMO was a Fortune 500 “tech-focused” company (?) which had offices scattered across the middle of the US.

Salaries. The book says that it was not in Silicon Valley or New York; the salaries were described as good, but in this context that meant “average over $90,000”. That is significantly less than the starting salary of US$120,000 for my fresh-out-of-coop SE students working in the US (per my2 019 exit survey), and we’re comparing new-grad salaries with salaries across all experience levels. Even if we accept that the US is an inequitable place, the starting salary in Canada for my students is still US$84k. TOMO employees are likely in lower cost of living places. I don’t think they come out ahead.

Position of IT. I think part of the issue is that IT was a cost center and not a profit center at TOMO, though I can’t get that for sure from the book. (What does “tech-focused” mean exactly? Is this company IBM-like?) Executives are always interested in reducing costs of cost centers. Hence significant downsizing and outsourcing. This is the sort of thing that drove CS enrollments down in the late 2000s. I guess downsizing + outsourcing, as feared by students, indeed arrived at some workplaces, though not all.

Buyer’s market. Even for my students, finding full-time jobs is stressful. Yet they manage. And I think that changing jobs is often lucrative for them. People at TOMO were worried about losing their jobs. They also had pride in what they’d accomplished earlier. OK, I sort of understand, no one wants to lose their job. But still. (And after the merger lots of people moved on, the book says.)

Process. TOMO was using a “high ceremony” process with lots of documentation and a pretty waterfall-esque software process, e.g. testers coming in after the developers did their thing. Definitely no DevOps: one handed off the project to operations after “finishing” it. They had also outsourced lots of development to India and paid bottom dollar for extremely junior developers there. As a result, workers had lots of meetings around the clock. What’s more, people were expected to be constantly available to answer questions from their managers that just weren’t that important.

What I didn’t remember reading about was “endless JIRA” that occurs in some workplaces. Maybe it’s too waterfall-y for that.

EDIT 20 Jan 2021: Kropp et al wrote a related paper, “Satisfaction and its correlates in agile software development”. In the surveyed Swiss companies, 80% of IT professionals reported at least somewhat Agile development, with the most plan-driven professionals reporting the most unsatisfaction with the methodology.

Compared to academia, and Waterloo in particular

I feel very privileged to work as faculty at the University of Waterloo. It’s by no means perfect. But from my privileged perspective this university gets many things right. I’ll focus on topics that are related to this book.

Here’s a workload survey across academia:

https://twitter.com/JessicaCalarco/status/1350141312587198466

(Was 2020 a great year or a terrible year for a sabbatical? For me, it worked out very well due to being in NZ and having no caregiving responsibilities. It’s like, would you rather be doing emergency remote teaching or not?)

Outsourcing

The equivalent to outsourcing for academia is adjunctification. We don’t physically outsource but we do conceptually outsource to low-paid, low-status, highly-qualified instructors. It’s driven by lack of money and desire for flexibility (you have to keep tenured faculty for a long time). But my department, Electrical & Computer Engineering, aims to stay below 10% classes taught by adjuncts.

Micro-management

Professors generally have a lot of control over their timetables; the department (or the Registrar’s Office) may impose a teaching schedule, but in-person instruction only accounted for 100 working hours of the year. Meetings are a few more hours.

Waterloo, as with many Canadian universities, has had a fairly health-conscious response to the pandemic, and not tried to open campus prematurely. I think it’s been about right.

We are empowered to teach and supervise graduate students as we judge best, with support available if we want it. As a concrete example, I submit my marks and that’s it; I don’t need to show them to anyone for approval. There is some oversight for undergraduate teaching: different departments have different practices, but Engineering programs solicit comments from student representatives through the term and pass them on to instructors (with varying levels of assertiveness; as SE Director I collected feedback and used my judgment in passing it on).

Reporting

For tenured faculty, every two years we submit a report on our activities in the past two years: teaching, research, and service. The department’s merit committee considers the reports and assigns performance numbers, which are linearly correlated with our salary increases. The numbers go up to Faculty level and can be changed at that level sometimes. Thus, the assignment of the numbers can sometimes be opaque but it’s generally reasonably fair.

One change that’s happened in my time at Waterloo is that we moved from every-year to every-two-years reviews for tenured faculty. Less paperwork!

COVID

That’s the multi-billion-dollar question. What’s going to happen to work post-COVID? I’ve heard that software companies are generally not keen to pay Bay Area salaries to people who are not living in the Bay Area. At the same time, I understand that many of my students have gone remote, since offices are closed.

The real estate trends have been for people to move away from downtowns in North America. I can understand that during a pandemic. The need to reduce commutes doesn’t quite exist, and the benefits of being downtown are non-existent. But I personally still wouldn’t want to live in a suburb. In the end we decided that returning to Canada during a pandemic had no real upsides: it would always be boring or terrifying, and living in a downtown isn’t great, and living in a suburb is always terrible. We’ll see what others do.

Are managers going to force their employees to come in? I think that the pandemic has been a big reset and there will at least be discussions of the use of offices. I’ve heard that some companies have cut down on planned office space acquisition.

I’ll be observing trends here with interest.

Conclusion

I’ve thought that work in 2021 sounds dysfunctional in some ways. This book points out exactly how it has been dysfunctional. But I don’t know which segments of the population it really affects. Once again, I’m very curious to see what peoples' experiences are. I encourage you to get in touch with me and tell me what you think!