# Help finding some monthly site stats about topics and user trust levels

**URL:** <https://meta.discourse.org/t/help-finding-some-monthly-site-stats-about-topics-and-user-trust-levels/42560>\
**Category:** Data & reporting\
**Tags:** sql-query\
**Created:** [2016年四月14日 15:04 UTC](https://meta.discourse.org/t/help-finding-some-monthly-site-stats-about-topics-and-user-trust-levels/42560 "2016-04-14T15:04:48Z")\
**Posts on this page:** 1\
**Showing post:** 11

<div class="post-metadata">

**Author:** ![DeanMarkTaylor](https://sea3.discourse-cdn.com/meta/user_avatar/meta.discourse.org/deanmarktaylor/32/102462_2.png) [@DeanMarkTaylor](https://meta.discourse.org/u/DeanMarkTaylor)\
**Post date:** [2016年四月14日 23:21 UTC](https://meta.discourse.org/t/help-finding-some-monthly-site-stats-about-topics-and-user-trust-levels/42560/11 "2016-04-14T23:21:13Z")

</div>

> [@tobiaseigen](#):
>
> so maybe my second idea of just tracking the number of posts every month to public community categories is more sensible/meaningful.

> [@DeanMarkTaylor](#):
>
> Now includes `reply_within_x_days` parameter to allow the topic to be counted if replies are within that period.

I believe as long as you don’t “use” last months figures until at least `reply_within_x_days` after the last day of the month these are good numbers.

Actually I’m seeing `1` days (78.6%), `4` days (78.6%) and `15` days (79.5%) for one sample set of data - so you could pick a smaller number of days and measure that, Yes `1` day and `4` days are the same.

---

_[View the full topic](https://meta.discourse.org/t/help-finding-some-monthly-site-stats-about-topics-and-user-trust-levels/42560)._