It's a great idea, and I've used this before for temperature logging in a different context, but in this case, I would like to have an apparatus I can send elsewhere for the data to be collected on their end without too many hassles as it's already a big ask for the temp to be monitored like this.
The idea is to monitor variation in temperature of stored biological materials; these are in thermostatically controlled refrigerators, and therefore there is temperature change at some small level as the thermostat kicks on and off. It is likely to be quite small- small enough 0.01C won't give fine resolution, so 0.001C would be better.
Because these products are stored differently- some in open refrigeration, like (for example) a can of soda on a refrigerator shelf, some in boxes, some in other forms of overpack, and in varying density- there will be differences in variation. IF (and that's a big "if") damage is caused by these tiny variations, then samples stored after years or decades may benefit from improved thermostatic regulation: insulated overpacks, for example, to reduce temperature fluctuations.
But first we need to see how much difference there is. So, for example, 1 second data points at 0.001C resolution at multiple sites would be of benefit. Automating this with datalogging would be key.
-AJ