How much does it cost to run an academic lab?

August 07, 2012

When I tell academic colleagues that I’m interested in crowdfunding my research, I hear the same skeptical refrain: “A couple of grand is peanuts compared to a R01, so what’s the point?” Obviously, to think we can catapult in a single bound from the $100s to $1,000s raised by recent science crowdfunding successes, e.g., the SciFund Challenge or whole-genome sequencing of rare disease patients, to the $100,000s is absurd. That’s why I’m planning a proof-of-concept crowdfunding experiment in the intermediate $10,000s range.


Which gets me thinking: on average, how much does it actually cost to run a small academic lab? There’s not much out there in the blogosphere in terms of budget datasets – if there are, please provide links below in the comment thread – so I decided to create my own. I’ve been a Lewis-Sigler Fellow at Princeton since 2007, where I’ve managed a $200,000/year budget and overseen a two-employee lab for the last 5 years.


By releasing these figures into the wild, I hope to stimulate several conversations, first and foremost, regarding efficiency. To clarify, I don’t exclusively mean preventing waste arising from ill-advised equipment purchases. Rather, I mean tackling the asymmetric marketplace in research transactions, along the lines of what science startups like Science Exchange and Assay Depot are doing. Second, if crowdfunding is ever going to compete with the government monopoly on funding, the global research community needs to share their spending habits in order to expose non-obvious inefficiencies, and to illuminate unmarked paths toward greater efficiency that become apparent when comparisons are made across many labs, regions and disciplines.


Okay, with that preamble out of the way, let’s talk numbers.


Each month, I receive an electronic statement detailing expenditures – in the parlance of commerce, my “burn rate.” Here I graphed my lab’s monthly burn rate over the length of my soon-to-be-completed appointment at Princeton:

I’ve color-coded fiscal years (July 1 – June 30) to make comparisons easier. Aside from a spike in September 2009, when I made a one-time > $40,000 instrument purchase, my burn rate has been fairly consistent. The single most expensive line item is – you guessed it – personnel, which consumed well over half of my budget. I should note that my lab is atypical in that I’ve employed two full-time Masters-level research associates (technicians). Experienced technicians are more expensive than trainees, e.g., graduate students and postdocs. Based on these numbers, one can begin to understand why the pitched cries over protracted apprenticeships as a form of indentured servitude have a basis in reality.


Interestingly, there appear to be lulls when my lab spent less money than average, and, conversely, volatile periods when my lab spent more money than average. To see if there are seasonal patterns, I graphed the burn rate data by month:

The Fall is the most volatile season, followed by a stretch of below-average consumption in the Winter and Spring, and concluding with a modest ramp up in the Summer. The Fall volatility is mostly driven by an artifact of my appointment start date. I attribute the under-consumption of Winter and Spring to thrift and my desire to maintain reserves, even though my funds roll over. However, by May it was usually clear how much money remained, and I became bolder with purchases in the Summer, not to mention that Summer is the academic conference season, and these meetings are expensive.


In Part 2, I examine my lab’s spending patterns at higher temporal resolution, and take a closer look at the costs of personnel vs. consumables/reagents/services.

Related Posts

  • Bill

    We could save a lot of money if there were an efficient comparison-shopping method or service for consumables, too — like Biocompare, except that actually worked. Maybe it exists and I just haven’t found it.

    There was also once a neat idea (with a strange name) called Bioroot — it’s dead now, because scientists hate to share — that tried to save money on expensive reagents by creating a central lab inventory. The idea was that you could e.g. call someone up and ask for a tiny amount of something like a spendy antibody, much less than the minimum purchase, when you just want to do a “try and see” experiment. Many a lab has scrapped an idea, or spent $500 on a one-off, when there was an unused tube of the same reagent just 5 minutes across campus.

    • William Gunn

      This is less relevant to life science than chemistry, but is pretty good for chemicals, and of course I’m a big fan of Science Exchange for helping to address some of the larger structural issues.

    • Ethan Perlstein

      I alluded to two scientific research-specific transactional marketplaces in the post above, Science Exchange and Assay Depot – basically Craigslists for experiments. (Full disclosure: I commissioned a project on Science Exchange, which is still ongoing).

      I don’t use Biocompare because there’s no incentive. My home university gets an institutional discount from Fisher, which is the Wal-Mart of laboratory supplies/reagents, offering everything imaginable via third-party vendors. Seems to me you’d have to address the supplier oligopoly if you want real price competition. But that’s just consumables.

      My local molecular biology department sends out emails to their list for “try and see” reagent requests and it seems to work well. I’ve always assumed this kind of informal, peer-to-peer reagent sharing happens at all departments but maybe that’s wrong.

  • Tetyana

    The inefficiencies in academic science (and I can’t speak for much else, my experience is limited) make me kind of crazy. I don’t think I will ever fund basic research, frankly, which sounds terrible, but, knowing how money is often spent (and my opinions are obviously based on my limited experiences)… I think particularly in some super-well funded labs… I don’t know, is buying pre-made agarose gels for everyday use like colony PCR really necessary?

    But more to the point, I have little idea how much it costs to run the lab I’m currently in, and I know that there are differences in the way funding works in the US and Canada, making comparisons harder (I would assume). But one thing that I always thought would make things SO much more efficient (and again, maybe I’m wrong, maybe what seems “obvious” is wrong) is having core facilities. For example, I’m in a worm lab. There are not that many worm labs in general, but my University has quite a fair number. Wouldn’t it be easier if there was a core facility set-up for even just the every-day simple things like plate pouring, seeding plates, freezing and thawing strains? (I wish that we had a facility similar to the Drosophila Media core-facility in Princeton, but for worms).

    It seems that good core-facilities really make scientists and researchers much happier (anecdotal experience) – why not increase them and increase the range of services they provide?

    Do scientists not like sharing because people are afraid of contamination, mix-ups? It seems like such an obvious way to save money..

    • Ethan Perlstein

      Tetyana, all sounds points in my opinion. It would be interesting to see survey data on how labs use core facilities. I’m sure such inquiries have been made by individual departments, but have their results been shared on the Internet stage for others to see? My guess is not likely.

  • Joanne Kamens

    And don’t forget about the large and growing open sharing community at, the mission driven non-profit plasmid repository. Full disclosure–I am the Exec Director at Addgene. We have over 1,300 different labs sharing almost 20,000 unique plasmid reagents all over the world. Depositing is free and there are small fees to request. EAch exchange is accompanied by a completed MTA via our innovative electronic MTA process (usually takes less than 2 days). We shipped over 60,000 plasmids in 2011. We love to work with people who are interested in open access and collaboration.

    • Ethan Perlstein

      Thanks, Joanne! I ordered plasmids from Addgene a few years ago and had a good experience.

  • Bonzo

    Very interesting pie chart. Minor graphic gripe – benefits (health insurance, etc. yes?) should be next to salaries & similar in hue – emphasizing the ~50% budget cost of people. ~25% miscellaneous? would be interesting to see what this comprises.

  • Nathaniel

    This is really interesting, thanks for posting!

  • Genecks

    I like this article. One of the things that consistently bothers me about scientific research is what is really holding back science? From this page, I would come to believe that what is holding back science is actually the cost to fund employees. However, there are individuals like me who enjoy scientific research, being in a lab, and the idea of discovering new things and furthering the realm of science: I like neuroscience.

    It surprised me when I saw that there was a tiny slice when it came to buying research supplies. However, I am not sure what the costs involved of “special purpose” equipment involves: Pipettes? Spectra equipment?

    I can tell you, however, from being an undergraduate student at the University of Illinois at Chicago (B.S. Neuroscience;2011), that what I continually noticed about many of the biology and medical science labs is that people left around 5 p.m.. I remember talking to my research manager/professor about the efficiencies of UIC, and I recall that he stated that one of the reasons UIC is not a big, top school is people go home on weekends… However, I did a lot of math, and I consider that even if people did work on weekends, such as the Ph.D graduate students, they would perhaps shave off 6 months of time from their Ph.D length.

    I understand that there exist bottlenecks in scientific research: Such as taking an hour or so for a gel to complete. Various experiments have bottlenecks where you’re sitting around wishing results would arrive in the snap of your fingers. However, that is not what happens. Nonetheless, I believe if people were TRULY scientists, they would be hanging around the laboratories a lot more, getting more volunteers into the lab, and progressing the number of scientists who are actively attempting to further research and discoveries. However, that’s not what is going on: At least, that’s not what was going on at UIC, for what I saw. I did see some people busting their arses on the weekend, but those people were few. Those people were serious workers, people with a serious work ethic, and I was definitely a researcher who stayed late and worked on weekends. I enjoyed being in a laboratory.

    The pay is low; but I’ve considered over a lot of time that what would help science progress is if people went to jobs that made them more money than the science job. Many people could be working other jobs, such as nursing, physician’s assistant, chemist, engineer, etc… Many times because the science jobs are volatile and can let up easily. And from what I’ve learned and studied, having a lot of research scientists, even on a volunteer basis, continually in the lab, allows the progress to move forward without missing a beat.

    It’s like the days of the hardcore scientist are gone and replaced with money mongrols who could care less about the field and more concerned about finances. However, even for those individuals, with a continued influx of passionate scientists, labs would stay active in the death of the night until the 9 to 5’ers show up.

    I have yet to really understand what it is. I think some of these “bigshot” research professors throw in the towel too early and argue they have other obligations.

  • Genecks

    To say the least, a decently setup scientific protocol to conduct experiments in order to get results should enable a highschool student to walk in and add to science. I think the inability of research labs to have people constantly working on research really reflects on the incompetency of those who manage the labs. No offense. I know this, because when I was first hired into a lab, the instructions I was given were ridiculous. It talked about making EDTA solution, yet lacked the proper methodology for going about it. I sometimes wonder if people keep these things as a riddle (as trade secrets) so that science becomes bottlenecked for the gain of those who keep to themselves knowledge of how to properly go about particular experiments.