5
\$\begingroup\$

In our recent visual excerpt of a year in moderation, we see that community moderation actions have drastically declined. One explanation is that our site-wide quality improved drastically — simultaneously, our site analytics are in decline. What is going on here?

\$\endgroup\$
7
  • 2
    \$\begingroup\$ Can you explain what you mean by "in decline"? \$\endgroup\$
    – VLAZ
    Feb 4, 2023 at 14:37
  • 3
    \$\begingroup\$ Analytics viewing requires 25k rep, although given you can see them I might be missing something. Can you cite some of the relevant analytics for those of us who cannot see? \$\endgroup\$
    – ESCE
    Feb 5, 2023 at 7:32
  • 3
    \$\begingroup\$ Here's a screenshot of the analytics covering February 2022 to Jan 2023 inclusive, for those who can't access the analytics themselves. I'll leave it to Akixkisu to elaborate on how they're interpreting it for the question. \$\endgroup\$
    – Carcer
    Feb 5, 2023 at 13:40
  • 1
    \$\begingroup\$ @VLAZ Akixkisu might misidentify the may 2022 drop in reported traffic that comes from the cookie swap. \$\endgroup\$
    – Trish
    Feb 5, 2023 at 15:13
  • 1
    \$\begingroup\$ @Trish possible, but I still don't see how "traffic" it's related to "quality". The claim doesn't make sense even when I know about the cookie. \$\endgroup\$
    – VLAZ
    Feb 5, 2023 at 15:16
  • \$\begingroup\$ @VLAZ that is true, but it is most likely the source for the claim of site analytics in decline. \$\endgroup\$
    – Trish
    Feb 5, 2023 at 15:18
  • \$\begingroup\$ I think this is a general trend that started in 2019-2020 ish and should be looked at in that space of time — a general traffic trend over the total time trending upwards (in particular rising towards 2016) can be identified, but part of why I pose the question here in this state is that I want other people to look at the data without my interpretation of the data. \$\endgroup\$ Feb 6, 2023 at 0:04

1 Answer 1

11
\$\begingroup\$

Not necessarily.

The apparent drop in traffic is an artifact of how cookies for Google Analytics are categorized

On the 10th of May 2022, the cookies used by SE were changed. As a result Site traffic fell off the table the week of May 10, 2022. What happened?

Due to the altered cookies, pre-May 2022 data can't be reliably compared to post-May 2022 data.

The drop in "traffic" coincides with the swap of the cookies to a different model. Before that day, the analytics tracking cookie was mandatory, since then it is optional and positive acceptance is required to be tracked. As a result, tracked traffic declined since many people deny performance tracking cookies by default.

Handled questions are not correlating to quality of questions but quantity of handlers

The number of questions handled in some way or another does not only correlate with the quality of the questions, but the quantity of handlers. Watching the number of handling queue events handled by the site, a direct impact on all other handling categories above is to be expected.

Or phrased differently: Many people identify many bad questions. If fewer people identify fewer bad questions, that doesn't mean there are more good questions, but just that there are fewer bad questions detected.

\$\endgroup\$
2
  • \$\begingroup\$ Have you gone back further in time and looked at the trend starting in 2019? \$\endgroup\$ Feb 5, 2023 at 23:54
  • \$\begingroup\$ @Akixkisu as said, I see a reduction in handling jobs. \$\endgroup\$
    – Trish
    Feb 6, 2023 at 0:04

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .