The Steam Workshop Attention Dilution (2026): More Mods, Less Attention Per Mod Than Ever

The Steam Workshop has never been bigger. It has also never been harder to get noticed.

Using ModSignal's live dataset of 7,259,401 active Workshop items, we tracked every mod's subscription count against the year it was published. The result is a decade-long trend that should reframe how anyone — creator, studio, or player — interprets the Workshop's headline growth number.

1.19 million mods were published in 2025 alone — seven times more than in 2015. But the average mod published in 2025 earned 816 subscriptions, down from 4,120 in 2015. That is not a slow decline. It is an 80% drop in per-mod attention, spread across a catalog that kept doubling.

The Workshop is not shrinking. It is diluting.

The decade-long slide

Grouping every active mod by the year it was published, and measuring what that cohort actually earned, produces a trend line that only points one way:

Two things are happening simultaneously, and they are not unrelated. The mod count keeps climbing — 2026 is already at 995K through September, on pace for ~1.33 million — while the median mod published in 2026 has 7 subscriptions and 8 lifetime views. That is a mod that was uploaded, seen by a handful of people, and passed by.

The view collapse is the more revealing number. Total cohort views peaked in 2022 at 1.69 billion for 692K mods. The 2025 cohort — with 1.19 million mods — received 1.20 billion views. Fewer total views spread across 72% more mods. The Workshop's discovery surface is not expanding at the same rate as its supply, and each new mod's slice of attention is shrinking as a result.

The shrinking success rate

The percentage of mods that reach 1,000 subscriptions — a rough threshold for "someone noticed this" — has fallen every year since 2019:

We limited this to January through September for an apples-to-apples comparison, since 2026 is not yet a full year. Even so, the trend is stark: a mod published in the first nine months of 2023 was 1.9 times more likely to reach 1,000 subscriptions than one published in the same period of 2026.

The absolute number of breakthrough mods is also falling. 2025 produced 110,103 mods at 1k+ across the full year — the highest ever. But 2026's Jan-Sep count of 62,450 puts it on pace for roughly 83,000 for the full year, which would be the first annual decline in breakthrough mods since 2022.

At the other end, the zero-subscription rate is climbing in the opposite direction — from 14.8% of 2015's mods to 19.2% of 2026's. The Workshop's bottom is growing faster than its top.

The dead zone

The most striking number in the dataset is not the top or the bottom. It is the band in between — the mods that exist, that are technically live, that were uploaded and published and visible, and that earned fewer than 10 subscriptions.

3,389,015 mods — 46.7% of the entire active Workshop — sit in this dead zone. Their combined subscriptions total 7,084,043, which is 0.056% of the platform's 12.5 billion total.

Nearly half the Workshop. Less than one-sixteenth of one percent of its attention.

The full subscription distribution makes the concentration explicit:

70.9% of all mods have fewer than 100 subscriptions and together hold 0.57% of the platform's subscriptions. The top 0.1% — 7,259 mods — hold 35.7% of every subscription on the Workshop. The top 1% hold 69.7%. The bottom 50% hold 0.1%.

This is not a long tail. A long tail implies a gradient — diminishing but present. The Workshop's distribution is a cliff: you are either above it or you are in a crowd of millions that collectively does not matter to the subscription economy.

Where the dilution concentrates

The attention dilution is not evenly spread. Some ecosystems are drowning in low-attention mods while others maintain a healthy ratio. Among games with the most mods published in 2025–2026:

Rust and Counter-Strike 2 are the dilution engines: 88% and 78% of their 2025–26 mods have zero subscriptions. Trailmakers is a different kind of problem — only 1.8% are zero-sub, but just 0.61% reach 1k, meaning 98% of its output sits in the 1–999 band that the platform's attention economy barely registers.

At the other end, Garry's Mod and Left 4 Dead 2 continue to produce breakthroughs at rates that look like a different platform entirely — 24.9% and 17.0% of their 2025–26 mods clearing 1,000. Wallpaper Engine is in its own category: 90.6% of its new mods reach 1k, because the wallpapers are consumed in-client, not browsed as Workshop pages.

The total attention ceiling

The most counterintuitive finding is that total Workshop subscriptions per cohort peaked in 2021 and have been declining since — even as the number of mods published per year keeps breaking records:

The 2021 cohort — 561K mods — collected 1.42 billion subscriptions. The 2025 cohort — 1.19 million mods — collected 971 million. More mods, fewer total subscriptions. The 2026 cohort is on pace for ~750 million, which would be the lowest cohort total since 2018.

There are two ways to read this. The optimistic read: older mods have had more time to accumulate subscriptions, so the gap will narrow as 2025–26 mods age. The structural read: the Workshop's total attention budget — the combined willingness of every player to subscribe to anything — has a ceiling, and supply has blown past it. The conversion rate (subs ÷ views) has stayed remarkably stable at ~80% across all cohorts, which means the bottleneck is not that players are less willing to subscribe. It is that fewer of them are seeing each mod in the first place.

What this means

Three things follow:

1. The headline mod count is a vanity metric. A game with 200K mods and a 40% zero-sub rate is not five times healthier than one with 40K mods and a 5% zero-sub rate. It is a larger pile with a smaller active core. The success rate — what fraction of mods actually reach an audience — is the metric that tracks ecosystem health.

2. The competitive bar is rising every year. A mod published in 2026 is competing against nearly a million other mods published in the same year, plus the entire accumulated back catalog. The same quality that would have reached 1,000 subscribers 70% of the time in 2019 now reaches them 6% of the time. Creators who succeeded in 2020 and returned in 2026 are operating in a fundamentally harder market.

3. The Workshop is approaching a quality-vs-quantity inflection. Total cohort subs have been declining for four straight years while mod count keeps rising. If the trend holds, 2026 will be the year where the Workshop published more mods than ever and collected fewer subscriptions than any cohort since 2018. The supply is still growing. The demand is not.

The Steam Workshop's real story in 2026 is not that it has 7.3 million mods. It is that 3.4 million of them — nearly half — are standing in a room where almost nobody is looking. The attention didn't disappear. It just spread too thin to notice them.