We didn't know what part of the virus would change before the virus started changing. A small number of patients developed broadly-neutralizing antibodies (bnAB

We didn't know what part of the virus would change before the virus started changing. A small number of patients developed broadly-neutralizing antibodies (bnABs), which not only wiped out Covid classic, but kept working well against the other variants too. Once we found out what part of the spike protein those antibodies bound to, then we knew what to target. But it's still not easy to raise antibodies to recognize specific sites on a protein - the more "context" you include, the more likely it is that your immune system will find another, less effective, handle on the protein to grab and call it a day. I suppose we knew about pan-coronavirus conserved sites before, but it (imo) would have been too risky to try targeting those sites with the first mRNA vaccines - one experimental technology at a time.