5 years after COVID: what did modellers get right and wrong?

The COVID-19 pandemic represented a major challenge to many sectors of society. It also provided the opportunity for epidemiological modellers to prove their worth. Much of the modelling was performed to extremely tight deadlines and was underpinned by noisy and often biased data.
5 years on, and with the benefit of hindsight, I’ll present a personal perspective of what went well, what went badly and lessons for next time. I’ll cover many aspects, but pay particular attention to vaccination, roadmaps, Omicron and building collaborative networks.