Team Science Best Practice #4 - Build strong research support systems
Post 6 in the Six Team Science Best Practices series
Research support systems, a category which includes project management, information management, and financial management, are absolutely critical for conducting science efficiently and effectively. Yet few of us receive any training in these areas during our PhD programs or when beginning to build our research programs. We are expected to develop and implement such systems in support of our research using existing university-provided tools that are rarely tailored to the complexities of scientific work. (Data management systems will be discussed in the next best practice post!)
Making the concept of building strong research support systems even more challenging is that, in full transparency, not a lot of research has been done in this area in pursuit of developing best practices that researchers can use. What works in a typical office setting may not work in a distributed research team in the university setting, where many team members are working on multiple projects and expected to follow different processes for each one.
I am currently leading an R01 funded by the National Library of Medicine to study information in translational research teams called Information Management Prototype for Clinical and Translational Research (IMPACT-CTR). If you are a funded team working in translational research (very broadly defined!) and would like to learn more about participating, please visit our study website and complete the form to have a brief informational call. The requirements for participation are minimal and you will be contributing to the development of best practices in information management!
In addition to making your research more efficient by helping you stay organized and on top of what’s going on with your team and your project, spending time early on in your career to build research support systems that work in your particular context has a couple of additional benefits. First, by investing in system building when your team is small, you can figure out what works for you and be ready to scale those systems as your team grows. Second, research systems can help boost your productivity without you simply working harder. Finally, if you have really good research systems that work for you, when you start collaborating with others, they may be willing to use your trusty and efficient approach, saving you the trouble of having to adopt someone else’s systems!
Different disciplines and different kinds of scientists (e.g., a bench scientist vs. a health services researcher) may look different on the surface, but there are enough commonalities that we can share some general principles of system development.
Principles of System Development
Budget for system development work. I cannot stress enough how important it is to allocate resources to engage in system development. If you have a start-up package, this is an ideal use of those funds, as strong research support systems will pay huge dividends in research productivity later on. I encourage everyone to budget for project management support, even if, on paper, it looks like you are then doing less science. Without these systems in place, you will be spending huge amounts of time doing project management instead of doing the science you love. The amount of time and money to budget for developing and supporting the systems, as well as potentially paying for the systems themselves, will vary widely by discipline, career and project stage, and project type. If you’re unsure, ask around to find people who are a few years further down the road than you are and ask what they wish they’d done differently! (This kind of work should absolutely be provided and supported by universities, but that is a rant for another day.)
Simple is best. Especially when you are just getting started with system development, start as simply as you can. There is no need to create an elaborate project management system, for example, if you are working on a small pilot study with one other person. Even as your team and projects grow in size and scope, simple systems that require minimal upkeep and user training will give you the best return on investment.
Take the time to do a very simple scope of work for each system. What do you need this system to do? What are the pain points you’re experiencing in your research (e.g., versioning of the protocol, can’t find the meeting minutes or decision records)? What platforms are available for you for free or very low cost? Note that this scope should include what all of the members of your team need to do their jobs. Some universities have IT or strategic planning experts who may be able to help with this step.
There is no “perfect” platform for any of this stuff. The right platform for your team is the one that requires the least amount of upkeep, is easy for people to use, and meets your scope of work. That could be google docs or something more advanced like SharePoint. A simple system that everyone understands and is willing to use in the way you want them to will always beat out a more complex system that is confusing and will spur work-arounds.
Conduct regular trainings in your systems and send your staff and trainees to trainings in the theories behind them. Training staff and trainees in using the systems you implement has several benefits. First, of course, is that they are more likely to use them and use them correctly if they understand the processes and your expectations! Second, teams change and evolve over time, so conducting regular trainings ensures that the new members are using the systems properly while also benefiting from the experience of more seasoned team members. Third, these trainings may surface challenges or needed improvements you hadn’t previously noticed and can be addressed. Finally, as discussed, research support systems are essential parts of doing science and if your staff and trainees don’t have those skills, your research will suffer. Furthermore, your staff and trainees need these skills for the future and by providing them with training (e.g., a class in project management techniques), you are contributing to their future career growth. Most universities have free or low-cost project management or data management courses for staff and trainees - take advantage of them!
Some flexibility will enhance your system. We each come to our teamwork with our individual systems in place that work (maybe!) for us but probably conflict with how others on our team work. Creating processes within your systems that allow a little flexibility for how your team members’ brains work can help your system be more successful. One way to consider these different approaches is to gather info during the development of your scope of work. How do team members currently approach their work and can you build ways for them to maintain the essence of those approaches while standardizing?
Trust but verify. Once you have everyone trained on using the system, check to make sure they are actually using it and using it the way you want them to. Repeat the training session every time someone new joins the team. I have seen so many projects go astray when team members decided to secretly use their own systems rather than the ones agreed to by the team. You will end up with lost information and even lost data this way, not to mention frustration.
Again, this work should be easier than it is. Funders should require these systems for every project and universities should make high-quality research support systems easy to identify, assess, develop, and implement for research teams. Without this institutional-level support, every research leader is forced to learn this skill set and build their own systems, without the benefit of broad expertise and communal learning. Further, funders should be supporting more research in this area so that we actually know what makes for good research support systems. Research support systems have a huge impact on how research gets done and we should demand the same kind of rigor here as we do with other aspects of research.