That being said, the process generally doesn't change — you still grow and kill (or grow and weaken) viruses the same way, you just start from a different base strain. For already-approved inactivated-virus flu vaccines, the FDA does not require new clinical trials for strain updates, and for already-approved attenuated virus vaccines, they only require very minimal trials (300 adults to prove adequate attenuation) [2]. So they're not novel in the sense of the COVID vaccines, which needed large scale clinical trials to prove efficacy — the flu vaccines have generally already proved efficacy, so for strain updates the most they need to prove is safety.
As I understand it predicting which flu strains to vaccinate this year is particularly hard because mask wearing/isolation/social distancing has reduced flu levels enough that they don't have a lot of data to base their predictions on
OK, this is the part I wasn't aware of. I assumed (dangerous, I know) that each year's 'new' flu vaccine was really just a new 'blend' of existing proven vaccines, for the most prevalent strains that year.
Edited to remove badly conceived question. New Question: How often do we have to develop novel vaccines for new Flu variants?