Let’s take the Moderna trial for example, which doses on Day 1 and Day 29. I can’t find specific published data, but safe to assume that “Day 29” is not absolut

Let’s take the Moderna trial for example, which doses on Day 1 and Day 29. I can’t find specific published data, but safe to assume that “Day 29” is not absolutely 28 days later in every case, but on average 28 days later. So the “fail safe” method would be to have a 100% reserve of second doses at all times. No first dose can be given if a second dose is not reserved. Alternatively, you give out all your doses as fast as they come in, and you have a separate queue for 2nd-dosers who get priority when they schedule their time slot. Some days your whole supply could go to 2nd shots. But unless you’re willing to actually delay the second dose for several months, the total number of people vaccinated in the first 3 months doesn’t really change all that much between the two scenarios. In either case you shouldn’t literally be locking up doses for 30 days. The smart thing to do logistically is you have tracking on doses throughout the delivery pipeline, probably knowing your specific scheduled deliveries at least 14 days out. The second dose for a person dosed today doesn’t have to sit in the freezer from Day 1, it just needs to be reserved from the delivery expected on Day 24. If a shipment gets behind and someone gets their second shot on Day 30 or 31, obviously this is perfectly fine. The only issue with getting ahead of yourself would be a major unexpected supply chain interruption that could leave people waiting a couple weeks longer, but then we have much bigger issues.