Starting your own research group is a daunting proposition. TIR offers some pointers to help you on your way…

Do it your way
There’s no “right” way to go about it, so don’t get yourself in a muddle by overthinking it. Just do it, learn from your mistakes, and try to take on board as much advice as seems relevant. Almost inevitably, you’ll probably find that you end up repeating all the mistakes that people warned you about, but that’s part of the rite of passage. Don’t beat yourself up about it.

Don’t stress about the finances
The most expensive thing is people, and people will probably be hard to come by. It’s obviously an unfamiliar situation to suddenly be in charge of a budget, but don’t worry about it. No lab failed through running out of money for consumables. 

Be thrifty, but don’t over-economise
The second bit of financial advice. Be wary of splurging large sums of money for hardware unless it’s absolutely essential and there isn’t something good enough within reach. Utilise the available infrastructure as much as possible, get a sense of what kind of things can and can’t be done easily. On the flip side, don’t scrimp with things for your personnel – their salaries are the single largest expense for the group, so you want them working at high efficiency. Get them the best items you can afford, and don’t say no to requests unless they’re unreasonable. You have to invest to be successful.

Maintain research focus (1)
Don’t fritter away energy and time by trying to look at too many things at once. Keep things focused on a single research area until you’re well-established.

Maintain research focus (2)
To start with, have everybody in the group working on ONE PAPER AT A TIME. Everybody works on that one project, then when it’s published, everybody switches to the next paper. This can be a very hard lesson for postdocs that have come from an environment with a high level of independence and operational autonomy – it naturally feels hypocritical to be corralling everybody, but if you’re using undergraduates this is almost essential. Every paper that’s published will be a boost to the group’s morale. Once you’ve published a bit and maybe have a few more people with you, then you can think about having two projects on the go simultaneously within the group. One caveat – don’t commit to authorships too early as this can be a source of conflict; ensure authorship order is determined democratically.

Figure out what you need
Think what the lab’s core techniques are (e.g immunoblotting, immunofluorescence microscopy, flow cytometry, whatever). Then list absolutely every reagent and bit of hardware you need to do those things. Then make sure you have them. Make a list of all the basic labware and solutions that you had as a postdoc, decide which things you think you still need, and acquire them. 

Digitise your protocols
It’s time-consuming in the first instance, but pays off in spades later on. Get all your protocols typed up, even the trivial stuff, so that it can be distributed amongst lab members and incoming students. Keep updating the protocols, but save each update as a new version so you have an archive of progressive optimisation. Put all the protocols on a lab Dropbox folder or similar so that your longer-term group members can access them without having to ask you.

Publishing
Don’t go all-out for one big multi-year publication. Try to publish good quality work as often as possible so that you establish yourself as a senior author. The breakthrough papers can come later. Do however try and find out from your departmental mentor what kind of publication objectives are expected of you. To some extent this will depend on the kind of infrastructure you have access to.

Write a review
If possible, try to write a review in the first year or two that the field will appreciate. This is an easy way of announcing yourself, and a good way of getting on top of (or refreshing) the theoretical background. Don’t write a book chapter – nobody will cite it.

Have a career Plan B
You want to be able to focus on the science, not stress about whether you’ll still have a lab in 2/3/5/whatever years’ time. That said, you will almost certainly have at least one career near-death experience along the way. Paradoxically, a very good way of handling this is to have a well-researched career Plan B ready to hand. Ask yourself what you could do if you could no longer remain in academia as a group leader. If you would like to stay in science, then what? Industry, publishing, communication, teaching, administration? Something else? E-mail people to get advice on what the careers are like on the inside. Knowing that you have options elsewhere is a great way of mitigating stress. You won’t feel so helpless during the pressure periods.

Hire cautiously
It’s better to have no people than to have bad people. Make sure new people coming in the lab are enthusiastic and that you like them. Get a sense of how much lab experience they already have. Get them to talk about their previous work to see if they understood what they were doing. Write a complete project outline for them (bigger than the duration of the project itself) before they start so that they can see how their bit of work fits into a larger picture.

Hire cautiously (2)
Don’t grow the group too quickly. You’ll probably only be able to directly supervise 2-3 people anyway, so don’t enlarge beyond 4 people until you’re ready and able to delegate some of the supervision.

Develop an interview technique
Ask people soft questions to put them at ease (“Did you find the way ok?”), let them talk (“Why did you apply?”), ask questions of what they did previously, then tell them what you do. Make sure candidates talk to any other members of the group, and listen to their feedback afterwards (you want to avoid dominant negatives). Be alert to possible cultural problems or things that may make it hard for them to integrate. Trust your gut feelings.

Supervise carefully
Most students will need a fairly high level of oversight to maintain full productivity, and keeping track of multiple people can be difficult. Start a timeline-type document for each student and update it with what they’ve been doing that week. Ideally meet once a week to discuss results and set targets for the next week, and make it clear that you are always available for small queries. Help and encourage your students – this is a real “pay it forward” exercise that will give you a reputation as a good mentor.

Don’t rush to leave the bench behind
To begin with, you will be the lab postdoc. You have the most experience, the highest standards, and the most know-how. Don’t lose that impact by spending all day on the computer!

Quiet time
Make sure you have some kind of protected time for reading and thinking. Whether it’s a train commute, a day working from home, or a fixed chunk of time during working day. You need to keep on top of the literature, and you need to be able to maintain a perspective on what the group is doing. This is impossible if you start drowning in minutiae.

Don’t martyr yourself
Keep a work/life balance. The name of the game is efficiency, not input. Working hard is not the same as working long hours. Encourage the same attitude in your students.

Make sure there’s a lab manager
The exact organisation may vary from place to place. You may need or have a technician who can double up as a lab manager, or it may be that a technician or secretary belonging to the department handles that aspect of things. Either way, you need someone who can handle the purchases and make buffers – this is not something you want to be committing time to yourself. Give people in the group lab jobs to maintain the infrastructure.

Be a good colleague
Interact with the other members of the department, try to be helpful, and don’t rock the boat too much if there are things you’d like to see changed. Your career will almost certainly depend on people’s goodwill at some point, so it’s easiest if you don’t generate friction. Get advice from your head of department in terms of strategy, and what they expect of you. Try and build a network of people you feel comfortable talking to, both scientifically and in terms of emotional support.

Don’t avoid teaching
Too high a teaching load is crippling, but an absence of teaching isn’t helpful either – it’s the best way of building a profile amongst the students, who are your recruitment pool. Again, work on a phased teaching plan with your boss that lets you get some experience without compromising your research commitment overmuch. Teaching experience is also vitally helpful when you’re applying for other jobs further on (if you make it that far).

Go on the lecture circuit
Hustle for informal seminars as much as possible, identify departments which are in your area and try to get there on a visit. This helps announce you to the local community and could potentially stimulate collaborations. Plus, these are the people that will probably be reviewing your papers and grants, so it doesn’t hurt to establish some kind of personal rapport.

Have a buddy
Without a doubt, you will have moments – if not days or weeks – when you feel utterly lost and alone and overwhelmed by it all. This is normal. Ideally, try and find a peer or someone who’s on the same track as you but a couple of years ahead who you can talk to and share with. If there’s nobody local, make contact with some of your PhD or postdoc peers and arrange a Skype-with-drinks session.

Be resilient
It’s tough, no two ways about it. You will probably be in an unfamiliar environment, trying to get things set up without prior experience, and surrounded by a new set of colleagues. You will probably have repeated episodes where you just feel miserable and lost. You will find that paper reviews probably become harsher now that you no longer have your postdoc mentor on the author list to protect you. You will get your grants rejected. You will have to deal with all this, over and over again, and with little encouragement from outside. Bouncing back, and developing ways of handling stress and existential angst is essential. Hang in there – we’re all in this together.
Acknowledgements/disclaimer
TIR is extremely grateful to the following people, who were basically the ghost writers of this posting: Claudine Kraft (U. Freiburg, Germany), Tim Levine (UCL, UK), Sascha Martens (MFPL, Austria), Tom Misteli (NIH, USA), Laurence Pelletier (U. Toronto, Canada), James Poulet (MDC, Germany), Karin Römisch (U. Saarland), Steve Royle (U. Warwick, UK), Joachim Seeman (U. Texas Southwestern, USA), Jim Shorter (U. Pennsylvania, USA), Alex Stark (IMP, Austria), Graham Warren (LMCB, UK). 

As usual, it’s worth stressing that you can ignore everything above and still get it right, so think of the content as being things to consider rather than a cast-iron protocol that should be followed to the letter. For a more extended but extremely useful consideration of the same material, check out Kathy Barker’s “At the Helm”, and the Wellcome/HHMI’s “Making the right moves”. Both are packed with invaluable advice and recommendations. 

And good luck! Anything we forgot? Let us know.