What Is a UTM Link? Definition, Parameters and Examples

A UTM link is an ordinary URL with UTM parameters (small tags) added to its query string so that analytics tools can attribute the visit to a specific source, medium, and campaign. UTM stands for Urchin Tracking Module, after the analytics software that introduced the convention, and the tags are just extra name-value pairs after a question mark in the address, for example https://www.example.com/spring?utm_source=newsletter&utm_medium=email&utm_campaign=spring-sale. When someone clicks that link and lands on your page, the analytics tool reads the parameters and files the session under the source, medium, and campaign you named, so you can see, in your reports, that this visit came from the spring-sale email rather than being lumped into a vague bucket like 'direct' or 'other'. The link works exactly like the untagged version for the visitor: it opens the same page, with the same content. The tags change nothing about the page; they only label the visit for measurement. That labelling is the entire point, it lets you tell which channels, campaigns, and individual links actually bring visitors and conversions. The one condition is consistency: UTMs only produce clean reports if you name them the same way every time, because analytics treats every different spelling as a different source.

There are five UTM parameters, and three of them are the ones you use most. utm_source records where the traffic comes from, the specific site, platform, or publication, such as newsletter, google, or facebook. utm_medium records the type of channel, the general category, such as email, cpc (paid search), social, or referral. utm_campaign records the specific campaign or promotion the link belongs to, such as spring-sale, so you can group all the links for one push together. Those three are the core set and answer the practical questions: which platform, what kind of channel, and which campaign. The remaining two are optional and more specialised. utm_term records the paid keyword the link is tied to, used mainly in paid search to see which search terms drove the click. utm_content distinguishes between different links or versions that point to the same place, so it is used to tell apart two calls-to-action in the same email, or the two variants of an A/B test. In a URL they appear after a question mark, joined by ampersands, and their order does not matter.

A UTM link works by carrying its attribution labels in the URL itself, so no special tracking is needed at the destination beyond an analytics tool that already reads UTMs. The sequence is simple. You build the link by adding UTM parameters to the destination URL, so the campaign details travel with the address. You then publish that tagged link wherever the campaign lives, in an email, a social post, an ad, or a banner. When a visitor clicks it, their browser opens the destination page and passes the whole URL, parameters included, to that page. The analytics tool on the page reads the utm_source, utm_medium, utm_campaign, and any optional utm_term or utm_content values from the query string, and attributes the session to exactly those values. From then on, everything that visitor does in that session, the pages they view, whether they convert, how much they spend, is filed under that source, medium, and campaign in your reports. The result is that you can open your analytics and see, channel by channel and campaign by campaign, which links brought visitors and which of those visitors were worth the most, rather than guessing. The parameters do nothing to the page; they exist purely so the visit can be labelled and counted correctly.

Fragmented UTM reports are almost always a naming problem, and it is the single most common way UTM tracking fails. Analytics tools treat UTM values as literal text, so Email, email, and e-mail are three different mediums to them, and Newsletter and newsletter are two different sources. If different people, or the same person on different days, tag links inconsistently, your one email campaign gets split across several rows in the report, and no single row shows its true total. The traffic is all there, but it is scattered, so you underestimate every channel and cannot compare them fairly. The fix is discipline, not more tools. Agree one naming convention for your team and write it down: decide the exact spelling and case for every common source and medium, standardise on lowercase to remove a whole class of mistakes, and pick a consistent style for multi-word values (such as hyphens, spring-sale not Spring Sale). Use a link builder or a shared spreadsheet so links are created from the agreed values rather than typed from memory, and document the convention where everyone can find it. Get the naming right and the reports assemble themselves; get it wrong and no amount of analysis can un-scramble the data after the fact.

No, you should not put UTM tags on internal links, links that point from one page of your site to another page of the same site, and doing so is a classic mistake that corrupts your data. The reason is how analytics handles a new UTM-tagged URL: when a visitor clicks a link carrying UTM parameters, most analytics tools start a new session and re-attribute it to whatever source and medium the tags say. So if you UTM-tag a link on your own homepage that points to your product page, a visitor who clicks it has their session restarted and re-credited to that internal campaign, wiping out the real source that actually brought them to your site, the organic search, the newsletter, the ad. You lose the true origin of the visit and inflate an internal 'campaign' that never brought anyone from outside. UTM links are for external traffic: emails, ads, social posts, partner sites, anywhere you are driving people to your site from somewhere else. To track how visitors move between pages inside your site, use your analytics tool's built-in behaviour, funnel, and event reports, which are designed for that, rather than bending UTMs to a job they break.

A UTM link and a tracking pixel both help attribute activity, but they work in opposite directions and answer different questions. A UTM link labels an inbound click: it puts the campaign details into the URL so that when someone arrives on your site, your analytics can tell where they came from. It lives in the address of a link you publish, and it needs an analytics tool on the destination page to read it. A tracking pixel or tracking code, by contrast, sits on a page (or in an email) and fires a signal back to a server when that page or email is loaded, recording that an event happened, a page view, an email open, a conversion, often for advertising or remarketing. Put simply: a UTM link tags the journey in, so you know the origin of a visit; a pixel or tracking code sits on the page and records what happens once someone is there. They are complementary. A common setup uses UTM links to attribute the click to the right campaign, and a tracking code on the landing page to record whether that visitor then converted, so the two together tell you not just where traffic came from but what it was worth.

Omniconvert Explore is an A/B testing and experimentation platform, and it can segment experiments by UTM source, medium, and campaign, which turns your UTM tagging into a testing tool rather than just a reporting one. Because Explore reads the same UTM parameters your analytics does, you can slice any experiment by the channel that brought the visitor: see how the newsletter audience converts versus paid-search visitors, or how a particular campaign's traffic behaves on a landing page. That matters because different traffic sources behave differently, a visitor from a branded email arrives with more intent than a cold visitor from a display ad, so a page or offer that wins for one source can lose for another. With UTM-based segmentation you can go further than measuring: you can build and test a landing experience per source, tuning the message to match where the visitor came from, and let Explore report which variation wins for that segment with proper statistical significance, and lift the conversion rate for each channel on its own terms. Across more than 70,000 experiments, with an average uplift of 23.2%, source-level segmentation is one of the levers Explore gives you.