Steam tag trend
Sokoban Too few
Releases, reviews written and modeled launch revenue for Sokoban on Steam, month by month (955 games with an estimate carry the tag).
Reviews written
+5%
3,234 in Jul–Sep 2026 vs 3,090 in Jul–Sep 2025
Paid releases
+93%
54 vs 28, same months (of 59 and 28 in all)
Upcoming games
312
median 14 followers (312 with a count); their total followers +161% in 30 days, over 5 games read at both ends (the largest weigh most)
Top contributor to the growth: He Who Watches, 137 reviews in Jul–Sep 2026 against 0 a year earlier — 95% of the tag's increase.
We compare the last 3 complete months with the same 3 months a year earlier, which takes out the season. Reviews are counted on the tag’s paid games only (free-to-play games have no monthly review history with us), so releases are counted the same way: paid releases. A tag needs at least 30 paid releases and 1,000 reviews in both periods, and no more than 50% of its releases in the two years free to play; otherwise it is “too few”. Rising: review volume up 20% or more, by at least as much as paid releases, and no more than 50% of that growth from a single game. Cooling: review volume down 20% or more. Anything else: Steady.
Reviews written are summed over 715 paid Sokoban games whose review history we re-read on or after Oct 1, 2026, so their last complete month is whole; 0 more, last read before that, are left out of every month alike (we re-read a game about once a week).
Our follower history starts Jul 13, 2026, so the 30-day change has nothing older to compare with. A game carries many tags and counts in each. Correlation only: a trend describes the past, it does not forecast.
Releases
Sokoban games that first went on sale each month (tracked Steam games).
Reviews written
Steam reviews posted each month on Sokoban's paid games of any age — the demand side. Free-to-play games have no monthly review history with us, so they are not counted.
Median revenue of the month's releases, 3 months in
Modeled: each paid release's estimated revenue at the last month boundary on or before 3 months after it first sold (2–3 months of sales), median per month, over the month's paid releases. Releases with no review yet have no history and are left out, so this is the median of reviewed launches and runs above a typical launch; each month shows its n of the paid releases. Through Jun 2026; younger months are not 3 months old yet.
Modeled estimate, not reported sales
Month by month
| Month | Releases | Paid | Median revenue, 3 mo | Reviews written |
|---|---|---|---|---|
| Sep 2026 | 20 | 19 | not yet | 1,016 |
| Aug 2026 | 24 | 22 | not yet | 991 |
| Jul 2026 | 15 | 13 | not yet | 1,227 |
| Jun 2026 | 16 | 15 | ≈$882n=7 of 15 | 1,280 |
| May 2026 | 19 | 16 | ≈$1.3Kn=12 of 16 | 1,103 |
| Apr 2026 | 12 | 10 | ≈$432n=8 of 10 | 973 |
| Mar 2026 | 22 | 20 | ≈$510n=14 of 20 | 1,141 |
| Feb 2026 | 11 | 11 | ≈$866n=8 of 11 | 1,096 |
| Jan 2026 | 14 | 13 | ≈$545n=11 of 13 | 1,652 |
| Dec 2025 | 12 | 12 | ≈$340n=10 of 12 | 1,220 |
| Nov 2025 | 13 | 11 | ≈$422n=7 of 11 | 930 |
| Oct 2025 | 9 | 9 | ≈$636n=7 of 9 | 977 |
| Sep 2025 | 11 | 11 | ≈$4.5Kn=9 of 11 | 877 |
| Aug 2025 | 7 | 7 | ≈$528n=7 of 7 | 926 |
| Jul 2025 | 10 | 10 | ≈$809n=8 of 10 | 1,287 |
| Jun 2025 | 10 | 7 | ≈$760n=5 of 7 | 1,023 |
| May 2025 | 13 | 13 | ≈$622n=10 of 13 | 1,140 |
| Apr 2025 | 33 | 30 | ≈$271n=21 of 30 | 1,422 |
| Mar 2025 | 20 | 20 | ≈$394n=16 of 20 | 1,119 |
| Feb 2025 | 14 | 11 | ≈$814n=8 of 11 | 753 |
| Jan 2025 | 11 | 10 | ≈$877n=8 of 10 | 1,016 |
| Dec 2024 | 8 | 8 | ≈$533n=6 of 8 | 1,085 |
| Nov 2024 | 21 | 19 | ≈$315n=11 of 19 | 911 |
| Oct 2024 | 17 | 16 | ≈$363n=12 of 16 | 756 |