Sunday, January 18, 2015

Are you a grasshopper or an ant? Risk taking in lab management

I have been spending a lot of time thinking about risk-taking behavior lately. Just a few months shy of two years I have six people in the lab and staring at their faces at lab meeting fills me with happiness and dread. I have made clear promises on the number of years I can support them, but they are all great and the last thing I want to do is to let them go in the middle of a promising project...hence the dread to keep the grant money flowing to keep them going if they need more time.

Before starting my lab I asked a lot of people for advice and put together a couple of collections of words of wisdom (here and here). The most frequent and consistent advice was to hire very carefully and take my time to hire the right people, which I followed religiously. I waited for the lab to be set up and I was more than a year in before I posted a job for my first postdoc and it took two rounds of interviews, a couple of thousand dollars in travel money and six months before I found the right fit. The other ones came very quickly, partially because grant money generated new jobs which needed to be filled.

While most of the advice I received was unanimous and made sense, there was one topic which was divisive: How to expand. Some people said: "Hire as much as you can, blow through your start-up and get as many hands on deck as possible to generate preliminary data for more grants", and others recommended the opposite: "Save your startup for a rainy day, plan very carefully and deliberately so that you don't exhaust your resources and become overwhelmed". So what to do? People have been equally successful with both strategies, but the "go all in" approach was more appealing to me so I ended up with a full lab in less then 2 years (I designed the lab space for 6 people and my ideal lab size is 6-10).

Am I right? Am I wrong? Will I die in the winter like a grasshopper? I don't think it's that simple. The more I do this, the more I realize that your lab is a projection of yourself. As you define the culture in the lab and assemble the combination of people you want, you also set the long term strategy. I am a very careful planner, but I am also not afraid of risk. I think a good proportion of risk can lead to higher pay offs and that you have to incorporate risk in your strategy. You accept a certain amount of risk if you invest in the stock market or if you purchase real estate in a transitional neighborhood. You have to able to roll the dice and see how it goes. In a small business you must include systemic losses in your business plan, i.e. that a percentage of your customers may not pay or that a certain amount of merchandise will be stolen. So my strategy is to have as many competent people as possible generate data in the best possible environment so that I will be able to write grants to support them further. Also, I like to collaborate widely and openly with the expectation that it's possible someone will screw me or take data from me, but that that is a risk worth taking in order to find people who will enrich my research instead.

I know people who are very uncomfortable with this approach, who got burned by past collaborators and do not want to share, or who want only 2-3 people in the lab to be able to work with them closely. And I think that's perfectly valid. What matters is that you have a strategy and that you figure out what is comfortable for you. We'll see how mine works out...


  1. I do not agree with the slow hiring strategy, especially when one can find plenty of competent people in today's bloated market. Of course that does not translate to being careless - one can still plan their budget carefully, hire very good people and get the lab's engine running ASAP. As you stated, it is important to get preliminary results so that you can start submitting grants soon. As it is, it takes ages to get an RO1 funded these days so why waste any more time by hiring slow?

    1. I should have explained what I meant better. The advice was not to rush into hiring for the sake of filling the lab. Hiring the wrong people at the beginning can set you back, causing loss of time and money and aggravation. I have multiple friends who were severely burned by bad hiring decisions at the beginning. Hiring very good people when you have an empty lab is tricky and it may not be possible for everyone...

    2. You did explain well in your post. And I agree with your reply above too. It is absolutely important to hire smart, but hire quickly. My main point was that there are so many good folks on the job market now that with proper guidance (mentoring committee comes in here) and knowing exactly what you want and being able to convey that to the candidates should result in a decent hire. For example, when hiring a post-doc as a new PI, if you want them to concentrate on a pet project of yours for the next three years, saying so explicitly during the interview stage and making sure the postdoc is okay with that and not totally wanting to go in their own direction can be the difference between a good hire and a terrible hire for a new lab. I feel, based on my limited experience, that it is a communication issue between PI and lab personnel that makes or breaks a lab.