Lethal Company
analyzed 2026-09-24 · sample of 621 English reviews across the whole review history
Multiplayer is the most frequently mentioned aspect, appearing in 17.1% of sampled reviews, and it is overwhelmingly positive there. Players describe messing with friends and dying together, with one writing "We all died a dozen or so times while beating each other with shovels until 2 am." Atmosphere draws praise in 5.0% of reviews, with comments on noises, monsters, and the feeling that something is out there. Mods come up positively in 3.1%, and price value is positive in every one of its 1.3% of mentions. Opinion is less settled on gameplay, which splits across 2.3% of reviews, and difficulty, where positive and negative mentions are even. Bugs appear in 1.4% and lean negative, with players questioning scanner behavior and trading quirks. Community and emotional impact each appear in 1.4% and lean positive, covering new friends, laughter, and terror. Developer conduct, content amount, and proximity chat also recur as things players bring up.
numbers checked against this report's stored data · quotes copied verbatim from reviews · plain text is the language model's own wording
Aspects
share of sample, axis to 20% · ± is the sampling uncertainty around each share
- positive
- mixed / neutral
- negative
-
multiplayer
95 positive · 5 negative · 6 mixed · 0 neutral
-
Amazing indie horror game with co-op, exploration, fun mechanics and good game play overall?
-
Can I mess with my friends and make them hate me? Yes
-
We all died a dozen or so times while beating each other with shovels until 2 am.
-
-
atmosphere
28 positive · 2 negative · 1 mixed · 0 neutral
-
Some good cheap thrills with monsters jumping out and such, and light comedic effects in some of the situations.
-
Great ambience, love the proximity chat and mods are easy to use
-
Proximity chat is the best part of this game but its still scary when you play it, all the noises and monsters also having the fear of knowing that something is out there is the best feeling.
-
-
mods
16 positive · 3 negative · 0 mixed · 0 neutral
-
Great ambience, love the proximity chat and mods are easy to use
-
You can use mods to boost the experience and make it better, its up to you.
-
run mods and this game turns in to a massacre (run with buddys)
-
-
gameplay
7 positive · 5 negative · 2 mixed · 0 neutral
-
Amazing indie horror game with co-op, exploration, fun mechanics and good game play overall?
-
Even in that role, however, this game is really fun.
-
Pretty simple job, lot of fun to be had in this little garbage man simulator so go out there, be a great asset to the company (the lethal company) and be a great assset, great great asset.
-
-
bugs
2 positive · 7 negative · 0 mixed · 0 neutral
-
Good enough for my thumbs up but buggy as hell
-
The game is great and all, but why does the manual say that only biological matter will glow red on the personal scanner, yet the turrets and landmines glow red?? Are they alive?
-
Some goblin wont trade my key for his pickle jar :<
-
-
community
7 positive · 2 negative · 0 mixed · 0 neutral
-
Public lobbies leads to new friends.
-
love it the people who play is very funny
-
Even without the endless mods the community makes for it, this game is still extremely fun!
-
-
emotional impact
7 positive · 2 negative · 0 mixed · 0 neutral
-
I have never felt this much terror in my life
-
This is by far the most i have ever laughed and also been the most scared in my life 11/11
-
first "horror" game that made me scream my lungs out so hard I coughed up blood
-
-
difficulty
3 positive · 3 negative · 2 mixed · 0 neutral
-
Is fun but it's hard. If you want real horror; single player.
-
If you want more of a horror game experience, play it solo but it is much harder that way and i dont recommend it.
-
i will say you will NEED to play this with some friends if you want it to not be hard as balls.
-
-
price value
8 positive · 0 negative · 0 mixed · 0 neutral
-
Best 10 bucks you can spend for hours of entertainment, what more could you ask for?
-
best 10 dolla game hands down.
-
It's very cheap, buy.
-
-
developer conduct
5 positive · 2 negative · 0 mixed · 0 neutral
-
… have been solved with a small tutorial, BUT other than that this is a wonderful game and would recommend to everyone and for a solo developer this is just outstanding.
-
Not to mention this was only made by one developer (who used to make Roblox maps).
-
i think the game is very fun and unique than other games and out sold call of duty which is funny and I do consider buying this game the shocking thing is that one guy made this game also my favorite monster is hoarding bug :)
-
see more
-
content amount
3 positive · 2 negative · 0 mixed · 0 neutral
-
very fun, add more quirky entities
-
the simplicity of the game is absoloutly ignored due to the randomly generated facilities, and a variety of creepy creatures.
-
Such a great game with so much content to explore with friends!
-
Emergent themes unvalidated
topics reviewers bring up that are not on the validated list; raw counts only, no error bars
- proximity chat 5
Review timeline
monthly peak 94,914 reviews · all languages
monthly share rated positive · 50% reference · months with <5 reviews faded
the timeline as a table
| month | reviews | up | down | positive | note |
|---|---|---|---|---|---|
| Oct 2023 | 1,331 | 1,311 | 20 | 98% | |
| Nov 2023 | 93,694 | 92,430 | 1,264 | 99% | |
| Dec 2023 | 94,914 | 92,383 | 2,531 | 97% | |
| Jan 2024 | 68,366 | 66,181 | 2,185 | 97% | |
| Feb 2024 | 29,208 | 28,113 | 1,095 | 96% | |
| Mar 2024 | 15,468 | 14,756 | 712 | 95% | |
| Apr 2024 | 16,411 | 15,691 | 720 | 96% | |
| May 2024 | 7,110 | 6,777 | 333 | 95% | |
| Jun 2024 | 8,376 | 8,011 | 365 | 96% | |
| Jul 2024 | 9,132 | 8,691 | 441 | 95% | |
| Aug 2024 | 8,620 | 8,206 | 414 | 95% | |
| Sep 2024 | 6,496 | 6,197 | 299 | 95% | |
| Oct 2024 | 5,945 | 5,655 | 290 | 95% | |
| Nov 2024 | 9,367 | 9,084 | 283 | 97% | |
| Dec 2024 | 8,297 | 8,042 | 255 | 97% | |
| Jan 2025 | 8,486 | 8,131 | 355 | 96% | |
| Feb 2025 | 5,887 | 5,656 | 231 | 96% | |
| Mar 2025 | 4,927 | 4,629 | 298 | 94% | |
| Apr 2025 | 2,454 | 2,276 | 178 | 93% | |
| May 2025 | 1,981 | 1,842 | 139 | 93% | |
| Jun 2025 | 3,121 | 2,956 | 165 | 95% | |
| Jul 2025 | 2,274 | 2,142 | 132 | 94% | |
| Aug 2025 | 1,910 | 1,744 | 166 | 91% | |
| Sep 2025 | 1,457 | 1,357 | 100 | 93% | |
| Oct 2025 | 1,897 | 1,733 | 164 | 91% | |
| Nov 2025 | 2,779 | 2,629 | 150 | 95% | |
| Dec 2025 | 2,047 | 1,934 | 113 | 94% | |
| Jan 2026 | 2,263 | 2,120 | 143 | 94% | |
| Feb 2026 | 1,694 | 1,590 | 104 | 94% | |
| Mar 2026 | 1,728 | 1,609 | 119 | 93% | |
| Apr 2026 | 2,751 | 2,592 | 159 | 94% | |
| May 2026 | 2,038 | 1,884 | 154 | 92% | |
| Jun 2026 | 1,579 | 1,460 | 119 | 92% | |
| Jul 2026 | 1,381 | 1,282 | 99 | 93% | |
| Aug 2026 | 1,260 | 1,165 | 95 | 92% | |
| Sep 2026 | 768 | 701 | 67 | 91% |
How this report was made
- Population
- all Steam reviews of the game (every language, purchase type, and verdict), walked newest-first per window; only English-language reviews are analyzed
- Sample
- 621 English reviews, time-proportional draw over 36 windows (newest-first within each), taken from an all-language walk and kept English-only — where English is a minority the sample runs smaller than the draw, and every margin of error is computed at 621
- Aspect yield
- 201 of 621 analyzed reviews carry at least one aspect (32.4%); every share on this page divides by all 621
- Interval regime
- calm: standard margins of error for a share of a sample (Wilson interval), no widening needed
- Fetch paths
- 36 time windows fetched directly (windowed)
- Language mix
- 1,000 fetched: english 762 · spanish 47 · russian 41 · 17 more
- Steam-marked windows
- none flagged by Steam (Valve's off-topic flagging began March 2019)
- Marked share
- 0.0% of sample inside marked windows, under the 2% floor
- Narrative
- passed the grounding gate first try
- Evidence floor
- aspects and emergent themes appear only with 5+ supporting reviews; smaller counts sit at the classifier's false-positive floor
- Instrument readings
- measured once on the study corpus for this model and prompt: frozen calibration, not properties of this run
- · aspect tagging vs. human labels
- 0.801 [0.752–0.844] (F1 against a human-labeled test set; 1.0 is best)
- · quotes filed under the wrong aspect
- 11.6% [6.6–19.6] of quotes, measured by hand on a sample; lower is better — measured on the prior model (2026-08-05); re-measurement pending
- · independent model vs. production labels
- 0.791 [0.772–0.810] (F1 of the production labeler against an independent judge model) — measured on the prior model (2026-07-25); re-measurement pending
- Versions
- deepseek-v4-flash · prompt classify-v1 · ontology v2
- Run
- serve-20260924T221326Z-2185084c · code 3eda721
full methodology in the milestone write-ups