What is a UTM parameter?

A UTM parameter is a short tag you add to the end of a link so your analytics tool knows where a visitor came from. UTM is usually spelled out as Urchin Tracking Module, after Urchin, the analytics software that grew into Google Analytics. A tagged link looks like this: yoursite.com/sale?utm_source=newsletter&utm_medium=email&utm_campaign=spring_sale. Everything after the question mark is the tag. The visitor sees the same page either way. Tags work on any link you control: ads, emails, social posts, QR codes on flyers, and links you hand to partners or influencers.

Without tags, a lot of your traffic lands in vague buckets like Direct or Referral, or in GA4's catch-all called Unassigned. Google defines Unassigned as "the value Analytics uses when there are no other channel rules that match the event data." Good tags are the cheapest way to shrink that bucket.

Which UTM tags should you use?

Google's campaign URL help page lists nine UTM parameters, but says you should always use three of them: utm_source, utm_medium and utm_campaign. Here is the question each common tag answers:

  • utm_source: who sent the visitor. Examples: google, facebook, newsletter, a partner's name.
  • utm_medium: what kind of traffic it is. Examples: cpc (a paid click), email, social, affiliate.
  • utm_campaign: which promotion. Example: spring_sale_2026.
  • utm_content: which ad or link, when one campaign has several. Example: red_banner vs. blue_banner.
  • utm_term: the paid keyword, mostly used for search ads.
  • utm_id: a campaign ID, useful if you later upload cost data into your analytics.
One campaign, five spellings: how 1,000 visits can split in a report
  • facebook420
  • Facebook260
  • fb150
  • FB110
  • facebook.com60

This is an illustrative calculation, not measured data: 1,000 visits from one campaign, tagged five ways, appear as five separate sources.

Source: Worked example

Why do small spelling differences split your reports?

Google states that UTM values are case sensitive: utm_source=google and utm_source=Google show up as two different sources. So when one teammate types Facebook, another types facebook and a third types fb, one campaign becomes three rows. Each row looks smaller than the truth, and nobody adds them back together.

Medium matters even more, because GA4 uses it to sort traffic into channels. As of October 2026, GA4's default channel group lists 19 channels, from Paid Search to Email to a newer AI Assistant channel. The Email channel only catches a medium of email, e-mail, e_mail or e mail. Tag a newsletter with utm_medium=newsletter and it misses the Email channel entirely. The chart below shows how quickly one campaign can fragment.

A naming convention you can actually keep

The fix is boring, and that is the point. Write the rules down once and ask everyone who builds links to follow them.

  • Use lowercase only. Pick facebook, not Facebook or FB.
  • Use underscores instead of spaces, and skip special characters.
  • Keep a short approved list for source and medium, using GA4's standard medium words such as cpc, email, social, affiliate and referral.
  • Name campaigns as what_when, such as spring_sale_2026, so they sort neatly.
  • Log every tagged link in one shared sheet with the date, owner and destination.
  • Never tag links between pages of your own site. A tag on a homepage banner overwrites the real source of that visit.

What UTM tags can't tell you

Google's own Campaign URL Builder is enough to start, and Matrix ROI includes a campaign link builder that applies the same pattern for every channel or client. But tags only tell you where a click came from. They don't tell you whether the sale was real, refunded, or would have happened anyway.

Google Ads also adds its own click ID, called GCLID, through auto-tagging, and Google notes that in GA4 auto-tagging is prioritized over manual tagging. So for Google Ads traffic, your hand-made tags may be ignored when both are present.

That is why tags work best as one half of a pair. The other half is your own record of payments. Compare tagged visits with what actually landed in Stripe or your store, a process covered in how to reconcile ad data with your sales records. When the numbers disagree, ad platforms over-counting conversions is usually part of the story. Matrix ROI, built by Smith App Studio, puts what each platform claims beside what your payments verify, day by day.