Posts
Comments
Hmm, I might be misunderstanding your intention, but I don't believe this is the correct question to ask, as it assumes that 'reading an article' ~= satisfaction with LessWrong.
A classic counter example here is click-bait, where 'visits' and 'time on page' don't correlate with 'user reported value.'
If we take a simple view of LessWrong as a 'product' then success is typically measured by retention: what % of users choose to return to the site each week following their first visit.
If we're seeing weekly retention for existing cohorts drop over time, it suggests that long-time LessWrong readers net-net value the content less.
Now, the prickly parts here are:
a) Users who've joined in the past year likely value the current post landscape more than the old post landscape. Given this, a retention 'win' for older cohorts would likely be a 'loss' for newer cohorts.
b) LessWrong's "purpose" is much less clearcut than that of a typical business/product.
I think it would be fun & productive to reflect on a 'north star metric' for LW, similar to the QALY for EA, but it's beyond the scope of this reply : )