What do Steam total, trusted and tracked UTM visits mean?

Steam's UTM visit columns describe different slices of tagged traffic.

Total visits count arrivals with UTM parameters;
trusted visits exclude bots and crawlers;
tracked visits require a logged-in user and exclude abnormal traffic.

Use the column that matches your question, and keep its name beside any percentage you calculate. Valve: UTM Analytics. Steam will show you outcomes such as wishlist addition and sales.

Start with the report, not the number

“Steam visits” is too vague for a campaign review. Write down whether you are reading a campaign's UTM rows or the broader store traffic report.

Valve's store traffic documentation describes where your game appears and which sources send people to its page. An impression means it was displayed; a visit is a page load. Those measures answer different questions from a tagged campaign report. Valve: Store and Platform Traffic Reporting.

For your next review, use this question-to-measure table:

Your question	Measure to inspect	What to avoid concluding
How much tagged traffic was recorded?	Total UTM visits	That every visit represents a different interested player
How much remained after traffic filtering?	Trusted visits	That “trusted” means likely to buy
How much traffic was eligible for recorded conversion analysis?	Tracked visits	That the rest of the audience had no value
What wishlist outcome was attributed to these links?	UTM wishlists	That this captures the campaign's entire effect

These are investigation prompts. A single column should not decide whether you keep or cut a channel.

One campaign, three different percentages

Suppose Your Game has the following fictional, finalized results for one tagged campaign:

Measure	Count
Total UTM visits	1,000
Trusted visits	700
Tracked visits	200
Attributed wishlists	30

Thirty divided by 1,000 is 3%. Thirty divided by 700 is approximately 4.3%. Thirty divided by 200 is 15%.

The arithmetic is valid in each case, but the denominator changes the meaning. Reporting “our conversion rate was 15%” without naming tracked visits invites someone to compare it with another campaign's 3% rate calculated from total visits.

Label the measure explicitly: “30 attributed wishlists per 200 tracked visits.” This is a ratio of reported counts, not a guarantee that 15% of distinct people converted. Do not multiply that percentage by the other 800 visits to estimate missing wishlists. That would assume the unobserved audience behaved like the observed audience.

What to check when the mix changes

If a campaign suddenly has fewer tracked visits relative to its total, compare it with its own previous setup. Did the placement, audience, destination or link flow change? Was the earlier comparison based on the same fields and filters?

Treat those as hypotheses to investigate. A change in the ratio is useful because it tells you where to look, not because it diagnoses the cause on its own.

Also inspect the export when a column is zero. An unavailable field and an observed zero need different interpretations. Do not calculate a rate with a zero denominator; record it as unavailable.