Heres a thought.
If BTC was spammed, and today's mempool count is supposed to be "normal", then does that mean the usage of BTC hasnt grown that much?
I'd say that transaction volume can be a misleading indicator for network growth. The overall trends should agree, but very large divergences (spam attacks, periods of temporarily high activity, anomalies) can occur.
I have no doubt the network has grown in the last several months, but if you look at the last 6 months,
mempool size has increased as much as 50-100x over some periods. That can't possibly represent legitimate organic growth, especially when you consider how empty mempools are today.