UTM parameters are short tags added to the end of a URL (such as ?utm_source=newsletter&utm_medium=email&utm_campaign=spring_sale) that tell your analytics tool where a visit came from.
They don't change the page; they label the click so you can credit traffic, sign-ups and sales to the right campaign.
UTM stands for Urchin Tracking Module. The tags come from Urchin, a San Diego analytics company that Google agreed to buy in March 2005 and turned into Google Analytics.
Other analytics and testing tools read the same tags.
How UTM tracking works
When someone clicks a tagged link, the analytics script on the landing page reads the parameters from the URL and stores them with that session.
Reports can then split visits, conversions and revenue by source, medium and campaign.
Google's official UTM documentation says you should always use utm_source, utm_medium and utm_campaign. The full set it lists:
| Parameter | What it records | Example value |
|---|---|---|
| utm_source | Who sent the traffic | google, newsletter4 |
| utm_medium | The type of channel | cpc, email, banner |
| utm_campaign | The campaign or promotion | spring_sale |
| utm_id | A campaign ID | spring_sale_2026 |
| utm_term | The paid keyword | running_shoes |
| utm_content | Which ad or link inside a campaign | video_ad, carousel_ad |
| utm_source_platform | The platform that sent the traffic | Search Ads 360 |
Google also lists utm_creative_format and utm_marketing_tactic, but notes they aren't currently reported in Google Analytics properties.
Values are case sensitive, so utm_source=Google and utm_source=google show up as two different sources. Pick lowercase and stick to it.
How UTM parameters show up in CRO and A/B testing
For conversion work, UTM tags do two jobs.
First, they tell you which traffic converts badly, which is often where the best test ideas are.
A paid social campaign landing on a page built for search visitors is a common example.
Our attribution guide covers how to credit those conversions.
Second, they let you aim a test or a personalized experience at one audience.
If your testing tool can target by query parameter, a rule like "utm_source contains facebook" puts only those visitors into the experiment.
That is the basis of ad message match tests and of traffic-source segmentation.
One caution: tagging internal links (a homepage banner, for example) with UTM parameters overwrites the original source of the visit, so a sale from a Google ad can end up credited to your own banner.
Keep UTM tags on links that point to your site from somewhere else.
Examples from the database
Several tests in LiftArchive changed the experience only for visitors from a specific campaign.
That kind of audience is what UTM tags or ad click IDs let a site pick out.
The sources don't say exactly how each audience was identified.
- Laura Geller Beauty auto-applied the discount for paid campaign visitors and showed a confirmation banner, instead of making them type the code. Visually reports +15.9% revenue from paid campaign traffic.
- Bonafide Health reminded Facebook ad visitors of their 20% off code in the cart and checkout. Visually reports +3.5% conversion rate and +5.5% profit per session.
- KlientBoost built custom pages for LemonStand's competitor PPC campaigns, matched to what those searchers were looking for, for a reported +47% conversion rate.
The two Visually results are vendor case studies that don't state sample size or significance, so read them as direction rather than a forecast.
@DaveDiederen
VisuallyVisually
KlientBoost
SplitBaseRelated terms
- GA4 for CRO: where UTM-tagged traffic shows up in reports and how to use it for test ideas.
- Attribution: how conversions get credited to channels, with UTM data as one input.
- Ad message match: tests that align the landing page with the ad that sent the click.
- Full CRO glossary