I'm struggling to see how that function could be a bottleneck, and if it were, memoization would easily get rid of it without the ridiculous limitation.
It could be better for sure, but that function has been there for almost as long as the project has been open-source if I recall correctly. It probably hasn't been touched since then because it wasn't a high priority.
It actually should be near time, at most some days old posts. You can also lookup by url and search for whole domain or "subfolder", see https://discu.eu/search/
Beside the newsletter, there's also a browser extension which shows the presence of discussions for the page you're currently viewing: https://discu.eu/extension and a series of bots for Mastodon and Twitter which post regular updates: https://discu.eu/social
How many new young customers do they have? Maybe more people than before think about getting a life insurance and for whatever reason the more "fragile" ones are getting one.
These are most likely group life policies where the employer pays the premiums for a base amount for all employees. Since there is no charge to the employee for the base coverage, most would accept the coverage.
If the employee base changed it could end up being unrepresentative of the broader population, but it's hard to imagine what that might be. Healthier people quit? Healthier people left Indiana? A competitor insurance company signed all the healthier groups?
Big employers in the US offer group life insurance to their employees without the need to opt in. Of course they have opt in add ons to increase the coverage, but in any case if you are employed and you die (knock wood) the insurance company’s phone will ring.
