As laboratory operations prepare to restart, researchers will have to adjust to institutional distancing restrictions (see Nature 582, 15–16; 2020). Unfortunately, there’s no playbook. We need to set up our environment so that a distributed scientific workforce can thrive.

There’s an opportunity here to fix those parts of the research machine that are broken — including inefficiencies, redundancies and irreproducibilities. We shall need infrastructure that enables us to share data, to connect team members and to communicate asynchronously as well as synchronously. Despite a surge in software-as-a-service start-ups, there is no single solution to help a lab to achieve these goals. Instead, each will have to find the right set of tools for its operation.

Imagine scientists sharing data as they are generated, recording analyses as they happen, receiving feedback from multiple sources and making measurable progress on lab projects — all in separate shifts. Imagine now that this scientific nirvana persists beyond shift-work restrictions. The new way of operating will accelerate scientific progress — including towards therapeutics for COVID-19, which provoked it in the first place.