“The Lean PhD: Radically Improve the Efficiency, Quality and Impact of Your Research” by Dr. Julian Kirchherr, published 2018 as part of the Palgrave Research Skills Series, is a pragmatic, no-excuses roadmap for those who want to do better PhDs but don’t know where to begin. In light of the sobering reality that most PhDs are low output, low impact, and drag on too long, Kirchherr’s rational response is to reject the status-quo approach to academia and branch out instead to the tech-boom start-up mindset. Moving fast and breaking things is not only one hell of a way to make money, but, claims Kirchherr, to do away with the inefficiencies of the PhD experience. As much as I am overall impressed with the idea, I do, however, worry that taking this approach could degrade a lot of what academia stands for…

A Roadmap For Efficiency

Kirchherr’s main argument for similarity is that both the PhD and the start-up are typically one-person shows where you put everything on the line to create, from scratch, something totally novel on the bare-bones back of limited resources. It’s an intense and all-consuming process with no guarantee of success. 60% of start ups, he claims (90% claims google), go bust, and 50% of PhD students will drop out. Of those that remain to complete their theses, how many of them go on to find success in academia? According to Kirchherr, not enough to rationally justify the time, energy, resources, and risk.

One-third of doctorate recipients report no firm employment upon graduation and 40 percent of those with a doctorate eventually end up in non-research roles (one reason may be that 80 percent of postdocs earn less than the average construction worker). (95)

For every 200 people who complete a PhD, only seven will get a permanent academic post and only one will become a professor. We are suffering a glut of PhDs who cannot find academic jobs. (102)

How then do we — i.e., those of us who want a career in research — beat the odds? Kirchherr suggests we simply be so good they can’t ignore us.

Because the academic job market is brutal, you must excel in your PhD if you want to stand a chance. (103)

To be that good, Kirchherr suggests we study from the start-ups that succeeded against even more drastic odds. He outlines three core principles of these effective start-ups, and how to practically implement them in each stage of thesis development from designing a project, writing the papers, and graduating with the final document. These keys to success are: (1) End User Orientation, (2) Minimum Viable Product, and (3) Rapid Prototyping. Having now been introduced to them, I may never go back to seeing work any other way…

Business plans rarely survive contact with first customers. (33)

End User Orientation: While the PhD always was and always will be much of a personal endeavour judged, in our hearts, by how we feel about ourselves, at the end of the day it is also externally assessed. Who then are these external assessors, what do they want, and how can we best give it to them? Kirchherr suggests we keep this front of mind at all times. On a basic level, this means not writing a 10,000 word paper if the target journal word limit is 6,000 words. It might also mean making small concessions to please your reviewers. To write a paper for acceptance means being aware of what the journal wants, what the field expects, and where science is moving as a whole.

The lean start-up approach is all about ‘failing fast’. You build an MVP because you can then collect feedback from your end user as quickly as possible. If the end user does not embrace your idea after several rounds of rapid prototyping, you fail this idea, but you do not quit. (90)

Minimum Viable Product (MVP): A minimum viable product is the least completed version of the item required for an end-user to provide feedback. The MVP should, in itself, be functional, but would lack the bells and whistles and fancy features of the final product. This MVP is then the unit of iteration allowing you to do the minimal work for maximal feedback from your end user; loop and repeat, as the product moves towards completion.

The core reason for this writer’s block is the belief that anything written by a PhD student must be of the quality of a PhD. This is simply not the case. Only the final thesis needs to be of PhD quality, and one will reach this quality by producing drafts and iterating on them again and again. (57)

Rapid prototyping: Prototyping is what we do with the MVP, once we have it. You’ve isolated the core concepts, developed the minimal structure, and now you need to go out into the world and find out if the idea is any good. For the PhD, our main iterators are, of course, our supervisors. But there are many other people who can serve in that role such as co-author collaborators, fellow students, tangential academics, and even partners and housemates. The best commentary, however, will come from whichever of your iterators best resembles your end-user. In this case, the journal editorial team and peer reviewers.

The crux of the entire Lean PhD approach, therefore, lies in finding good iterators, and a great deal of the beginning of the book is spent outlining how to select a good PhD project, good university, and good supervisors. His advice may be sound, but it isn’t mind-blowing: find a completable project in an under-researched growth area at a top ranked university with an early/mid career ‘hungry’ supervisor who has plenty of time to spend coaching you… (Great, thanks, and for finances shall I find the pot of gold at the end of the rainbow too?) Failing that, he suggests expanding the iteration pool to conferences (he attended 20 — evidently he did find that pot of gold after all). Or iterating directly with the journal, who are not just end-user-proxies, but the end-users themselves.

[S]tudents wait too long to collect the valuable anonymous feedback from scholars that can guide their work. Lean methodologies suggest that a paper must be submitted to a journal sooner rather than later. After all, the peer reviewer of a journal is the end user of the execution phase of the PhD and thus at the core of iteration. The earlier you submit your paper to a journal, the sooner you will receive the most valuable feedback to further improve it. (49)

All together, following the edicts of these three concepts lead to a PhD approach revolutionarily different from the stereotypical image of hunched-over academics locked in dusty ivory towers grinding out impenetrable manifestos on obscure topics nobody wants. Instead the Lean PhD operates right on the bright light edge, taking minimal steps forward, iterating with end-user-proxies, taking another minimal step forward, iterating again, and so forth; remaining flexible, remaining innovative, remaining totally and completely on-track with what our iterators want. The Lean PhD disciple will never be stuck in a rut or backed into a corner, unable to complete their PhD or publish their research because, thanks to dynamic collaboration at every step of the way, they would have ensured what they were selling had a guaranteed buyer.

After reading his book and listening to his talk, I have a lot of respect for Kirchherr (who, by the way, finished his thesis in 21 months, published 6 papers and concurrently consulted for 3 external projects, before landing a faculty position directly after graduating). He seems like a hard-working, intelligent, and eminently reasonable guy and the advice he has communicated in this book is his best attempt at making his positive experience accessible for all. However, while I praise Kirchherr, I can’t quite praise all of the book.

The Death of Depth?

As much as I like the dynamic approach of the Lean PhD, I am concerned about where it might lead if this approach is optimised over traditional academia. In the hands of experienced academics with plenty of researcher integrity, it may be a productivity boost, nothing more. In the hands of those trying to ‘get away’ with fast PhDs, I think it would be fuel to the fire, encouraging rapid, sloppy work.

Any day you can save on your PhD can be one where you will earn more in another job. (101)

While many start-up founders claim to want to change the world with their enterprise, survey data on start-up founders suggests that this is not their main motivation when launching a start-up. Rather, the most commonly named motivation is ‘the desire to build wealth.’ (74)

This second quote is left somewhat stand alone in text, not directly tying in with the PhD analogy but it leaves me ill at ease. As much as the entrepreneurial spirit of the start-up analogy appeals to me, there is a dark side to the start-up culture of selfishness, greed, and money-making obsession that Kirchherr doesn’t explicitly address but makes reference to. The remainder of this passage goes on to talk about how the PhD dissertation will never be perfect (true) so don’t waste time on making it as good as it can be (what??) - just make it minimal enough to pass. His whole idea of the Minimum Viable Dissertation and the Minimum Viable Paper are predicated on an approach of doing the bare minimum work to get you over the line so you can get out and get on with earning more money.

Indeed, the strategy revolves around the concept of the least publishable unit (LPU). The LPU concept refers to a publication strategy that produces papers containing the smallest amount of information necessary to generate a publication in a peer-reviewed journal. (52)

Taken literally, couldn’t this be construed as salami slicing? Why not optimise for the appropriately publishable unit (APU) instead?

In developing these barely-enough papers he suggests doing barely-enough work.

It is usually sufficient to closely read a book or an article’s abstract (and sometimes also the conclusion), while skimming the remainder of the article, to gauge what the piece is (or is not) about. Only very few articles are extremely close to the paper or thesis the PhD student will eventually write. These are the only ones that the PhD student needs to read most carefully. (54)

Read only as much as necessary; write as soon as possible. (54)

A recipe for misunderstanding your subject matter! Maybe the practiced academic with some qualitative sense of their unknown unknowns may be able to sense where to stop, but is not the young academic better served by reading too much than too little? When the goal is cultivation of a deep and thorough, multifaceted, transdisciplinary understanding, publishing prematurely is unlikely to lead to anything good. Maybe in Kirchherr’s social science realm of anti-dam movements there is room for mistakes, but what about fields where the margin for error is so much tighter? Does Kirchherr seriously want to be injected with a vaccine which the researcher did the bare minimum to create? Would he vote for a government who practiced Minimum Viable Diplomacy? I wouldn’t. When it comes to admin and hoop jumping, minimum viability might be fine, but if an idea is worth doing a PhD on, it is worth doing deeply. If we let the capitalist pressure for speed and “efficiency” encroach into academia, there will be nowhere left for depth.

We must focus on quality and impact for their own sake, or there will be nothing worthwhile left in the academy. (104)

He says… but then penalises much of what is commonly associated with quality.

Slowness Is Not A Dirty Word

[S]lowness is a central value in academia, and it is linked to quality. Indeed, the belief that quality is a function of time is omnipresent. The common belief is that the more time you invest, the greater your work’s quality will be. (84)

[T]aking a lot of time for something can also be an indication that one is not particularly good at it. (85)

Kirchherr rejects the commonplace conception that good things take time with the argument that slowness is often a thin veil behind which incompetence and unaccountability fester. While there are no doubt plenty of citable examples where that is true, I bet I can come up with just as many where it isn’t.

Academia is knowledge work, which lacks a definite relationship between input and output. It’s not possible to track words per minute or hours per day to infer the quality and impact of a researcher’s work. In fields where you can grind on a topic for a month then wake up, suddenly, on a Saturday morning with a complete conception of the solution in mind, Kirchherr and I agree that time is loose. But where Kirchherr penalises it, I find that the freedom to take my time encourages exploration.

The Lean PhD, in its chase for efficiency, leaves no room for the process of discovery. When working to a deadline, there is less capacity for side tangents, Blue Sky exploration, or mucking around and failing and then finding something else entirely. Only when unencumbered by clients and expectations and a red bottom-line can the academic afford to read a paper that might not be helpful, but then, sometime later, makes a connection to present work, and leads to breakthrough. For myself, it was only through the time-consuming process of carefully combing the methods sections of hundreds of papers that I discovered the core of my thesis. Slowness might not breed quality, but rushing is certainly antithetical to it.

Besides, I don’t even agree that the Lean PhD is all that fast anyway.

Why Not Just Clock Out At 5?

If the Lean PhD method really is as efficient and fast as it claims to be, it should require less - not more - strain than the traditional mode to complete. Kirchherr himself claims it to be true and I don’t doubt that for him and his exceptional mind being perfectly productive for the hours he is “on” is possible, but for the average student like me, taking this book literally would most certainly lead to burn out.

On average, a start-up entrepreneur invests 55 to 100 hours per week in their endeavor. (83)

“If you go into a co-working space on a Saturday afternoon, I can tell you which start-ups will succeed, without even knowing what they do. Being there on the weekend is a huge indicator of success, mostly because these companies don’t just happen—they happen because of really hard work.” ~ Marissa Mayer (23)

“[The former Yahoo CEO] believes that the secret to success is working 130 hours a week. This book does not advocate that you work 130 hours a week to complete your PhD. It also does not promise that you will complete your PhD by adopting the four-hour working week. Any PhD will always require many hours. However, the lean PhD will significantly reduce the hours needed. I estimate that it may halve the time, given that I submitted my PhD in 45% of the average time needed until submission at my school.” (23)

I feel this last statement is misleading depending how you measure time. If the traditional student were to work 38 hrs per week (and I actually mean ‘do work’, not ‘pretend to do work while gossiping and fooling around at the office’), 48 weeks of the year, across 3.5 years (42 months), they would work 6,384 total hours. As Kirchherr completed his PhD in only 21 months, to be more efficient than the traditional schedule in terms of work hours he would have had to work less than 76 hours per week. Surely feasible except for the repeated reference to 130 hour weeks and weekend work and grind. Even if we say the Lean PhD student is recommended to work a slothful 100 hours per week, to finish in 21 months, this would take over 9,000 hours - 16 hours per day, 6 days a week - or the equivalent of 4.5 years in the traditional work mode… He does acknowledge how unreasonable this is though.

Much empirical research indicates that productivity has less to do with the amount of hours we invest in a single working day, and more with the rest we have. (23)

To avoid burnout, Kirchherr recommends that these hours be broken up into intensive sprints with clearly defined timelines and intentional break periods between. Maybe this works for you, maybe it doesn’t. I don’t see, if the Lean PhD really is the recipe for streamlined success, why not just clock out at 5 and come back sharp at 8 the next day?

If the iteration method really does get better results, faster, can’t you trust the PhD to handle itself within its allocated 38 hour work week and use the rest of your waking hours for extracurriculars to - if we take a purely career-oriented view - develop other skills? While Kirchherr spent 21 months grinding on his PhD (and 3 consulting projects), the traditional student may take twice as long, but for fewer concentrated hours, and will do many MANY other things while they’re at it. Taking myself as an example, I am far from a thesis-optimised machine… Instead, I optimise for a well-rounded academic life: I tutor, supervise junior students, apply for grants, help others with their projects, write, learn new coding languages, read books from other disciplines, volunteer for societies, listen to academia podcasts, and do a dozen other little things that contribute to a rich tapestry of career-adjacent thinking and developing that I would not have time for if I was at work 76+ hours per week. (Not to mention hobbies, sport, family, and friends…)

Furthermore, Kirchherr’s mission to complete his PhD in record time seems entirely to miss the point. While Kirchherr in his lecture sounds like a well-rounded, well-developed, happy and healthy individual, he reads as a little obsessive. Everything is about the end goal. For me, a little epicurean perhaps, I want to be here, first hand, immediate, present, as the long tossed-over thought resolves itself into the logical how-what-and-why of understanding and I get that flash or slowly dawning light of insight. I want to do my PhD not just to have done it, but to be doing it. It is not a means to some other end, but the purpose in the first place. To be doing research, as my job! I’m already living my dream! In Kirchherr’s rush to be a ‘Dr.’ with a faculty position, it seems he forgot to stop and smell the roses. As he comes into the PhD with an already developed career behind him, this might be the optimal strategy, but for me personally, a little more time does not go amiss.

My Iteration: The Lean PhD v2

Thus, I propose to take the best part of Kirchherr’s philosophy, dial down the grind, inject some love of pure research, and come up with my next generation of the prototype. I will keep the parts about building a good team, failing fast, iterating often, and being mindful of the end user, but I will change the goals. Rather than working as hard as possible to finish as soon as tenable, instead I propose we utilise the Lean PhD method within our existing work schedules and use the new-found productivity to buy ourselves more time; then use that time to invest deeper into our project. We will not publish the minimum viable dissertation or the minimum viable paper, we will achieve that minimal dissertation/paper in half the time of other students, but then use the rest of our allotted hours to make it better, deeper, higher quality.

I am not claiming that Dr. Julian Kirchherr himself cut corners or was unscientific but I am concerned that, out of context, the Lean PhD could seem to be advocating for those practices. Kirchherr may well just be superhuman. He not only pulled off an incredible feat of research productivity while seemingly maintaining his mental health, but wrote a frank and funny book about it, and continues to be successful in his career. His roadmap, however, is likely not appropriate for the rest of us. Taken with a grain of salt, I would recommend this book to practically every graduate student and will be thinking about it for a long time to come.