Management of (big) research teams

Ideas to successfully manage a team of researchers, and what success means in this context.

Some academic research teams are standalone, usually embedded in university departments, while others are part of research institutes or other structures with shared goals. Most teams are rather small, with an average (for biology research groups in the UK in 2015, from this reference) of around 7 members including staff and students. Published analyses of productivity of research groups suggest that small to medium-sized groups are more productive (same reference as above).

However, groups of much bigger size exist. There are not many, but groups of 30, 40 or even 100 members, with one PI, exist. They might have competitive advantages given the right conditions, or maybe they are just the unavoidable statistical outliers of an imperfect funding system. In any case, because of their oddness, learning about the way they are run could be useful for the smaller groups. The basic premise behind this post is that a big research group that has survived as such for some time is probably efficient and noteworthy in the way it’s managed. Their experience can be useful for smaller groups and also for created structures like research institutes that are in several ways similar to big research groups.

In connection with this, the existence of very big research groups puts to the test the definition of their productivity. In particular, it seems difficult to measure the role of PIs as teachers and mentors, and the contribution that post-docs and PhD students make in careers other than academic research. Additionally, one can often see examples that high productivity needs more than scientific talent, and group culture for instance has a big role in how productive a group is in the long term.

Recruitment. Really big research groups first grow to an intermediate size because of one or several PI that are able to gain funding while they employ talented people who are also successful in bringing in more funding, supervise students and so on. It is very important how the recruitment formal and informal processes in the group work, how the PIs and the group members are able to identify and attract other researchers with great talent and team building skills, and also with a same shared research vision. I believe this is the single most important factor for the creation of a big group, i.e. a PI with a remarkable scientific vision that also happens to be good at recognising talent and attracting it. This is also critical for small groups that while remaining small survive for a long time. Other pieces of the puzzle might be not in place but this is always quite outstanding in the big groups I have looked into.

In-house support. All big groups seem to have dedicated support staff other than the support staff at departments, schools, university services, that the group manages to hire and maintain over the years. Roles include project management, scientific coordination, laboratory management, help with proposal writing, financial administration, engagement/impact, business development, and others. It’s difficult to say how important this is for a group, but big groups that have been in place for a long time all seem to have a highly profesional support structure within the group that typically comprises of around 8-10% of the total group staff. Small sized groups might take advantage of this strategy as well, and should consider to budget support staff in their proposals.

Organisation of research activity. The group is formally or informally a combination of several research teams, each with one or several team leaders. The team leaders coordinate research activity within the team and ideally also manage the career aspirations, training needs, etc. of team members. One group in the UK asks all team members to «make a one page summary every two months describing their work, goals, collaborations, problems, & tasks with times. A key part of the OPS is that list the time left in the group, match this against their tasks, career aspirations».

• Team meetings. Team meetings of one kind or another are usually weekly. Some groups have a meeting with the whole group weekly and another with only the smaller team also weekly. One to one meetings with the PIs or team leaders are also usually planned (every week, every two weeks, monthly… ) and sometimes upon request, although I think most groups / PIs in reality have only a guidance in place and usually attend requests as they happen.

• Shared knowledge. Additionally to group meetings where ongoing work is presented and discussed, most groups usually have some kind of shared folders, group server or intranet, with different levels of access, where papers, awarded and rejected grants, raw laboratory data, presentations, meeting notes, and so on, are available. If managed properly this can be a great tool to teach students how to write, analyse and present information up to a high level.

Building group culture. A group, big or small, that openly discusses career aspirations of its members, that has a collaborative environment, that teaches not only the science but also how to be professional, transparent and accountable, has all the chances to be most productive in the long term. Its members are productive while in the group and also maintain productive links after leaving and moving to other groups, industry, or other fields altogether.

• Solving disputes. Groups with a good culture have figured out how to best deal with problems between members. This depends on the PI personality and social skills, also on the team members’. The same protocols may not work well everywhere. In general, PIs that behave in a professional and transparent manner, that have empathy with personal life situations, that admit and correct mistakes, that do not let grudges or jealousy between team members unanswered, set an example that team members follow. Groups that for whatever reason don’t resolve disputes are bound to have mediocre collaboration within the group.

Collaboration with other research groups. Bigger groups seem to have more research partners in academia and in industry in particular. The ratio of partners per team member is probably not that different, but bigger groups have a critical mass that allows access to networking and industry leads that are more unusual to see in small groups.

Want more? There are many several good resources to learn about research group management. Here, to choose just one, I recommend a 1983 book ( High Output Management by Andy Grove) actually not focused at all at research groups, but at team management in general. This is an excellent book to learn about delegating, organising meetings, the critical importance of the team’s culture, leveraging one’s tasks to maximise the output of the team, monitoring/evaluating with indicators, and more. This summary is a good resource.

To finish. Should all teams follow the same best practice? Absolutely not. Your group might be better off doing some of these and some other different stuff. The aim of this post is to explain how most big groups do it and what could be useful to groups of all sizes.