How we research Reddit

Where the data comes from

Public Reddit archives back to 2005, plus our own collection for recent months. Together that's 3.8B posts and 27.7B comments, from Jun 2005 to Sep 2026. Removed posts stay in the archive, which is the only way to measure things like removal rates at all.

Archives have gaps, and we keep a log of the ones we know about. A study that touches one of those periods says so on the page. The clearest example: vote counts for July to October 2023 are frozen in the public archives, so we don't rank posts from that stretch by votes.

How we measure

• Every number in a study comes from a query against the archive. Nothing is estimated from a survey or copied from someone else's chart.
• We define the sample in plain words: which subreddits, which posts or comments, which dates, what was left out and why.
• We look for the boring explanation before the interesting one. If a rise could come from a keyboard, a data gap or a few busy subreddits, we check that first and show the check.
• Each study has a section on what the data can't tell you. Patterns across millions of posts say nothing reliable about one post or one person.

What we don't publish

We report on subreddits and on Reddit as a whole. We don't single out individual users in our research, and we don't publish usernames or quote people in a way that points back to them.

Using our charts and data

Charts and data files from our studies are free to reuse under CC BY 4.0. Credit KissMyPost and link to the study you used. Journalists and researchers who need a cut of the data we didn't publish can get in touch.

How to cite a study

Use the citation at the end of each study. The format is:

KissMyPost Research (2026). Title of the study. KissMyPost. https://kissmypost.com/research/…

Corrections

If you think a number is wrong, tell us. When we change a study, the page gets a new “Updated” date and a note saying what changed.