Every creator remembers their first mod on the Steam Workshop. This series has already measured it: the debut signal showed that a creator's first result predicts almost everything, and the catalog ladder showed the hit rate climbing from 8.59% on mod number one to 30.11% by the hundred-and-first. Both of those looked at how many mods a creator has made. Neither looked at the one thing that sits between any two of them: how long the gap was.
We took the live ModSignal dataset — 6,005,712 active Workshop items carrying at least one subscription, made by 1,687,348 creators across 392 games, holding 12.53 billion subscriptions — and for every item that was not a creator's first, we measured the exact number of days since their previous release. Then we measured how often it reached 1,000 subscribers.
The result is one of the cleanest relationships in the dataset, and then it becomes a more interesting one once you stop averaging across creators and start comparing each creator to themselves.
The shape of supply
First, a fact about how the Workshop actually gets filled. Of the 4,318,364 releases that have a measurable gap behind them:
| Gap since the creator's previous mod | Releases | Share |
|---|---|---|
| Under 1 day | 1,774,632 | 41.1% |
| 365 days or more | 262,358 | 6.1% |
Forty-one percent of all follow-up releases land the same day as the creator's last one. The Workshop is not a place where creators publish on a schedule; it is a place where a minority of creators publish in bursts and then a smaller minority publish at all. If you want the supply to look like a rhythm, the rhythm has to be manufactured by the creators who bother.
But the releases that do have a rhythm behind them tell a story the bursts do not.
The ladder
Bucketing every follow-up release by the gap since the previous one, and measuring how often it reached 1,000 subscribers:
| Gap since previous release | Releases | Avg subscriptions | Reached 1,000 subs |
|---|---|---|---|
| Under 30 days | 3,241,333 | 2,076 | 16.80% |
| 30–60 days | 248,495 | 3,300 | 19.64% |
| 60–90 days | 137,218 | 3,308 | 18.54% |
| 90–180 days | 224,631 | 3,027 | 17.03% |
| 180–365 days | 201,886 | 2,563 | 14.94% |
| 365 days or more | 264,800 | 2,583 | 12.71% |
The line turns over. A mod published within a month of the creator's last one lands 1,000 subscribers 16.80% of the time; a mod published after a year of silence lands it 12.71% of the time. In between there is a narrow peak at one to three months: mods that arrive on a roughly monthly-to-quarterly beat earn an average of 3,300 subscriptions against 2,076 for the same-day dump and 2,583 for the year-later return.
That is the pooled picture, and it would be a mistake to stop here. This series has run the same test often enough to know that a pooled ladder is usually composition wearing a costume: creators who publish on a monthly beat might simply be the creators who are good at it, and creators who go quiet for a year might be the ones who already gave up. The only way to separate the behavior from the person is to hold the person fixed and compare their own mods to their own mods.
The control that changes the number
Group every creator who has at least one release in each of two cadence bands, and ask which band won for that creator — comparing the creator's own hit rate in one band to their own hit rate in the other.
Comparing a year-plus gap against an active gap on the same creator (142,361 creators with mods on both sides):
| Same creator's releases | Reached 1,000 subs |
|---|---|
| After a gap under 90 days | 19.25% |
| After a gap of 365 days or more | 13.66% |
62.0% of those creators did better on the active side. Not a tie, not a coin flip — the same person, publishing into the same games, lands subscribers more often when their previous mod is recent than when it is a year old.
Tighten it to the band the pooled ladder liked best — the one-to-three-month beat — and compare it against the year-plus return for creators who did both (74,876 creators):
| Same creator's releases | Reached 1,000 subs |
|---|---|
| After a 7–30 day gap | 23.78% |
| After a gap of 365 days or more | 16.00% |
59.4% of those creators did better on the monthly beat. The direction is identical, the size is larger, and it survives every way we sliced it.
The single cleanest cut is the creators who actually run a catalogue (18,295 creators with 20 or more mods, both sides present): their releases after a gap under 90 days reach 1,000 subscribers 23.01% of the time, against 18.27% for their releases after a year of silence — 68.4% of them did better active. Restricting to their one-to-three-month releases against their year-plus releases (16,970 creators) gives 28.86% against 18.75%, with 64.1% winning on the beat.
This is not the same as saying "fast is good". It says something narrower and more useful: a creator's own recent mod is a live asset, and a year of silence drains it.
It is a rhythm, not a dump
The obvious objection is that "under 90 days" is doing a lot of work, and it contains the single largest group in the catalogue — the 41.1% fired out the same day as the creator's last one. If the finding were just publish faster, publish more, the same-day dump would be the champion. It is not.
Compare the two fast bands on the same creator, and they come out dead level: across 168,546 creators with releases in both windows, the 7-to-30-day beat won 50.5% of the decided cases — a coin flip. The one-to-three-month window is not better than the same-day rush; they are both "recent", and the difference between them belongs to the individual creator, not to the cadence.
The gap that carries the whole effect is the year. That is the number that costs you, and it costs you when the creator is held fixed, the game is held fixed, and the release's position in the catalogue is held fixed.
On position: the catalogue ladder already showed that later mods hit more often, so an active creator and a dormant one might simply be sitting at different points in their own catalogue. Restrict both sides to releases at position three or later in the creator's run, so neither side is padded with debuts, and the result is unmoved: 55,385 creators, 60.8% winning on the one-to-three-month beat, 24.95% against 16.45%. Restrict further to creators with at least three releases on each side and it holds at 64.2% (28.09% against 22.03%). The effect is not the ladder behind the ladder.
And it is not a drift in the data: the share of releases arriving less than 30 days after the previous one has been flat between 73% and 76% every year from 2018 to 2026, and the year-plus share has barely moved from 5.5% to 6.8%. Whatever this effect is, it has been constant for eight years.
The year-long return is not rare
The reason this matters more than a footnote is how common the dormancy is. Of the 704,979 creators with more than one mod, 212,955 — 30.2% — went silent for at least a full year and then published again. Nearly a third of multi-mod creators came back from the dead. Their return is a large, measurable, and mostly disappointing event.
You can see it in individual accounts. Spence! has 272 mods: 18 of the 18 releases in their one-to-three-month windows cleared 1,000 subscribers, against 3 of the 4 that followed a year of silence. 🔰SweetFox🔰 posted 14 of 14 in the monthly band above the threshold and 3 of 3 after their year gaps. leafzxg cleared 23 of 34 on the beat and 2 of 3 on the returns. And in the other direction: K'Braid has 10 releases on the one-to-three-month beat with a single hit, and four releases after year-long silences with none; legodude494 has 55 monthly-band releases with 6 hits and 3 post-silence releases with zero.
The pattern in those accounts is the same as the aggregate: the returns are not catastrophic, they are just reliably worse.
The reverse test: the silence is not a reaction
There is one alternative explanation that a within-creator test cannot rule out on its own. What if a creator goes quiet because their last mod flopped — the year-long gap is a symptom of a bad result, not a cause of the next one? If that were the mechanism, then a mod that missed 1,000 subscribers should be followed by a longer gap than a mod that hit.
It is the other way around. The mean gap to a creator's next release is 77.1 days after a mod that cleared 1,000 subscribers and 84.8 days after a mod that missed it — the rush, such as it is, follows a success, not a failure. (The medians flip, 4 days after a hit against 2 after a miss, because the distributions are dominated by same-day bursts on both sides.) The silence is not a sulk. It is a decision made before the result was known, and whatever caused it, the next mod pays for it.
Where the beat wins, and where it does not
The rhythm-versus-dormancy test is not universal, and the exceptions are worth listing.
Across 118 games with at least 300 releases on each side, the monthly beat beat the year-later return in 100 of them — 85% — pooling to 20.34% against 12.08%. The single most dramatic case is Workers & Resources: Soviet Republic, where the one-to-three-month releases reach 1,000 subscribers 74.52% of the time against 32.53% for the returns — a 2.3x gap in a game whose community lives on a steady modding beat. Cities: Skylines runs 58.17% against 16.59%, Ravenfield 57.45% against 21.30%, Transport Fever 2 59.25% against 23.29%.
In 18 games the direction flips. Some of those are rounding: Wallpaper Engine sits at 93.24% against 94.40% because everything succeeds there and there is no room to improve; Conan Exiles Enhanced is a tie at 33.24% against 33.33%; Poly Bridge 3 and Universe Sandbox are dead ecosystems where nothing reaches 1,000 subscribers on either side. But some are real reversals, and they deserve to be named: Half-Life 2 at 16.19% against 28.17%, ARK: Survival Evolved at 22.91% against 34.13%, Dota 2 at 23.22% against 27.41%, Team Fortress 2 at 9.13% against 11.45%, and VTOL VR at 5.69% against 7.75%. These are the older, more crowded ecosystems where the audience does not track releases at all and a year-old mod is competing with a decade of catalogue rather than with the creator's own last effort.
Leave out the one game that saturates everything and the shape narrows but never inverts: excluding Wallpaper Engine, the ladder runs 14.82% (under 30 days) / 18.03% (30–90 days) / 15.15% (90–365 days) / 12.02% (year-plus). The peak moves to the quarter — and it still tops out above both the dump and the return.
What it means
If you have gone quiet, the cost is not zero and it is not permanent. A creator's next mod after a year of silence reaches 1,000 subscribers about 5 to 7 points below the same creator's recent-work rate — 18.75% against 28.86% in the biggest catalogs, 16.45% against 24.95% at matched position. That is the cadence tax. The way to avoid it is not to publish more mods; it is to not let the gap run to a year. A quarterly release keeps the audience live. A same-day flood does not help — it lands exactly where a monthly beat does, no better — and it feeds the discovery dilution this series already documented in the one-sitting problem and the attention dilution.
If you are reading a "publish more, publish faster" statistic, check what it is comparing. The pooled ladder in this article, taken alone, says a monthly beat is worth 4 points over a same-day dump and 4 over a year-later return. Hold the creator fixed and the same-day dump stops mattering: it is a coin flip against the monthly beat. The only band that reliably underperforms inside a creator's own catalogue is the year of silence, and that is the one number here that is a decision rather than a personality.
Related reading: the catalog ladder on how far into the catalogue a mod sits, the debut signal on the first mod, freshness payoff on what updating a mod is worth, the churn economy on mods losing subscribers, the revival economy on old mods that suddenly moved, and the one-subscriber floor on the 654,035 mods that never got past their first install.
Method: 6,005,712 Steam Workshop items tracked by ModSignal as of 3 October 2026 across 392 games — every item that is is_active and carries at least one subscription, of which 867,994 (14.45%) have subscriptions_current >= 1000. "Gap" is the exact number of days between an item's time_created and the time_created of the creator's immediately previous item, computed with a window function ordered by (time_created, id); only items published after their creator's first have a gap, leaving 4,318,364 measurable releases. "Reached 1,000 subscribers" is a point-in-time threshold, not a tracked trajectory. The paired control groups items by creator_steam_id, keeps only creators with at least one item on each side of the comparison, and compares the creator's own hit rate on each side; ties are counted separately and the win percentage is quoted over decided cases only. The position-controlled variants restrict both sides to items at row_number() >= 3 in the creator's catalogue, so neither side includes a debut. Per-game tables use games with at least 300 measurable releases on each side. Wallpaper Engine is never excluded from a headline table; its exclusion is reported as a control and the "leave-out" ladder uses the four coarse bands. Counts are from a live production crawl and drift by a few dozen rows between reads; the creator totals in this article are read from the same snapshot as the item totals.