First of all, you should be aware that stability is an individual property of each antibody. In our experience, many antibodies are more stable as a serum. After purification, they may not very stable and precipitate. This is sometimes caused by the harsh elution conditions after a protein G separation. Some antibodies are sensitive to freezing. Therefore, antibodies should not be frozen repeatedly. One way to store antibodies is to add 50% of ultrapure glycerol and keep them at -20 °C.

Polyclonals are usually fairly robust as you have multiple different antibody species, not all of which will be negatively impacted by the elution process. Nevertheless you should neutralise the eluted Ab as fast as possible. If all Ab aliquots have been frozen and stored in the same way then logically they should exhibit the same performance. To store antibodies frozen, you need to freeze QUICKLY, definitely not a slowly in a -20. Small aliquots therefore help, as does using liquid nitrogen if you have it, or dry ice/ethanol or a -70 freezer.

I think you are far more likely to be seeing an issue with the lateral flow procedure. While the advice to use glycerol is sound in most situations, this is not necessarily a good option with your antibody, as you are using it for CAPTURE in lateral flow. Thus what you add to the antibody for storage purposes must not prevent binding to the lateral flow membrane.

I'd focus on the method of freezing, as required, but more likely the many potential variables in lateral flow, and hopefully you will start to get more consistent results.

I keep protein G purified polyclonal antibodies in phosphate buffered saline with 0.05% sodium azide at 4oC for 2-3 years without significant loss of activity. Your problem is most likely with the lateral flow set up.