TL;DR
Steam's UTM tracking undercounts your marketing because it only counts wishlists and purchases made within 72 hours of a tracked click, in the same logged-in browser. Demo downloads and playtest joins never count, and attribution breaks whenever a player switches browsers or opens the desktop client to search manually. Low numbers usually mean lost attribution, not a failed campaign.
The most likely cause: the 72-hour, same-browser attribution window
The most likely cause is the attribution window: Steam credits a wishlist or purchase to your link only when it happens within 72 hours of the tracked click.
A UTM link is a store page link with tracking tags added to the end (utm_source, utm_campaign), which tell Steamworks which ad, post, or creator sent each visitor.
An attribution window is the countdown after a click during which a conversion still gets credited to that click. Steam's is 72 hours; a player who clicks your ad on Monday and wishlists on Friday earned your campaign a conversion the report will never show.
Steam also reports tracked visits and trusted visits as separate columns, and only trusted visits count toward conversions:
- A tracked visit is any store page arrival Steam recorded through your tagged link.
- A trusted visit is one Steam judges came from a real, logged-in person rather than a bot or search crawler, so it is the smaller number, and the only one that can produce a counted conversion.
Beyond the window itself, attribution survives a same-browser login and the handoff when Steam's mobile banner opens the app, but it breaks the moment a player switches browsers or opens the desktop client and searches for your game manually, which is exactly how many players buy.
Valve publishes no universal capture rate. A widely cited practitioner case study from a major PC games retailer (Feb 2026) found Steam's UTM tool captured under 10% of traffic it could attribute through its own tracking. If that ratio held for your campaign, 1,000 visits your channels genuinely drove could surface as fewer than 100 in the UTM dashboard. Treat it as one team's measured result, not an industry rate, but it shows the direction and the plausible scale of the gap.
Symptom, cause, and fix
| Symptom | Likely cause | Fix |
|---|---|---|
| Conversions look low the day after a campaign | Conversions finalize after roughly a 4-day lag | Wait four days before reading the number |
| Conversions far below your ad platform's click count | 72-hour window expired, or the player converted in another browser or the desktop client | Read UTM conversions as a floor; compare channels against each other, not against click counts |
| Visits recorded, conversions near zero | Traffic is logged-out or filtered; only trusted visits convert | Compare trusted visits (not tracked visits) to conversions before judging the campaign |
| A demo campaign shows zero conversions | Demo downloads are never counted as base-game conversions | Judge demo pushes on trusted visits and your Steamworks demo stats instead |
| One channel reads as zero everywhere | UTM tags stripped before the store page loaded | Click your own posted link and confirm the utm_ parameters survive to the final URL |
Wait four days before reading conversion numbers. Visits in the UTM report update hourly, but conversions finalize after roughly a 4-day lag, so a day-one read of a launch campaign is structurally low even before any attribution loss.
Compare conversions against trusted visits, not ad clicks. Your ad platform counts every click; Steam only counts the subset that arrived logged in, stayed in that browser, and acted within 72 hours. A campaign whose 1,000 platform clicks show up as fewer than 100 UTM conversions has not necessarily failed; the two tools are measuring different populations, and the practitioner case study above found exactly that scale of gap.
Check that your tags survive the redirect chain. Link shorteners, social apps, and in-app browsers can rewrite links along the way. Open your own posted link and confirm the utm_source and utm_campaign values are still on the store page URL that finally loads; a stripped tag turns a whole channel into untracked traffic.
Use the same undercount to compare channels, not to size them. The losses (window expiry, browser switches, logged-out visits) hit every channel, so relative rankings between channels are far more trustworthy than any absolute count. Pick winners on relative UTM performance; never report the raw number as total marketing impact.
When a low number is correct, not a tracking failure
Some zeroes are by design, and chasing them wastes a week.
- Demo downloads are not counted as base-game UTM conversions, so a demo-focused campaign reporting zero is behaving exactly as documented.
- Steam Playtest conversions are not reported either; playtest campaigns are invisible to UTM by design.
- A DLC package that includes the base game IS counted, so a bundle promotion can look strong in the same report where a demo push looks dead.
- Very small campaigns produce single-digit conversion counts, where one browser switch or one expired window visibly moves the total; at that scale the report is noise, not signal.
Update Visibility Rounds will not fill the gap either: Steam grants 5 lifetime rounds, each capped at 30 days or 1M impressions, post-launch only, as a promotional mechanism for major updates rather than an attribution tool. They are covered separately in Steam Update Visibility Rounds explained.
The escalation path when a spend decision rides on the number
When you are allocating real budget between channels, escalate past the UTM dashboard. Start with the fundamentals of marketing attribution for Steam games, which covers the full measurement stack rather than one report. Then add your own layer: route every channel through links you control so you capture clicks before Steam does, and correlate wishlist-timestamp spikes in Steamworks against your campaign calendar for a directional read the 72-hour window cannot erase.
Steam's dashboard will only ever show the slice of conversions that stayed in one browser for 72 hours. Attribution connects ad spend, Steam link-tracking data, and wishlist/conversion numbers by channel, so a weekly check shows which channel is actually working rather than one blended number that hides it.
Sources: Valve documentation on UTM analytics and visibility rounds. A practitioner case study from a major PC games retailer's performance-marketing team (Feb 2026), which is one team's measured result rather than an industry rate. Published Steam marketing analytics guidance (2026).
