I think that capacity is as good a proxy as any for estimating the growth of the network. It's flawed as you point out, but I'm not aware of any other metrics we can use. Number of daily users / settlements would be better if we could determine these reliably, but alas that's impossible to determine (and that's a good thing).

That said, I often see people comparing Lightning capacity to the liquidity of other L2s. This is a false comparison IMO as it's a micropayment platform, not a trading platform like say the ETH L2s. There are tons of Lightning-enabled wallets, exchanges, nodes, vendors, and oddball services (like SN and podcasting 2.0). I use Lightning on a daily basis. But if you looked strictly at capacity, you'd come away thinking that Lightning's adoption is only marginally greater than that of Liquid's.