Conversion Rate: How to Calculate, Measure, and Improve It
Your conversion rate is the single number that answers whether your website is actually working. Traffic is vanity.…

Most campaigns report to a channel called “direct / none.” That label is not a real answer. It means your analytics has no idea where the visit came from — because the link you sent was bare. Learning how to use UTM parameters fixes this completely, and it takes about ten minutes to set up properly.
I’ve helped dozens of teams get this right. The failures I see follow a predictable pattern: the parameters are there, but they’re inconsistent, case-jumbled, or tagged on the wrong links. This guide covers the full picture — what each parameter does, how to build them, where they show up in GA4, and the mistakes that quietly wreck your data.
UTM stands for Urchin Tracking Module. Urchin was the analytics software Google acquired in 2005 and turned into Google Analytics. The name is dated; the mechanism is essential.
UTM parameters are short key-value strings you append to any URL. When someone clicks a tagged link, those strings pass to your analytics tool as session-level attributes. Your tool stores them and lets you filter, segment, and compare traffic by source, channel, campaign, and more.
A complete tagged URL looks like this:
https://example.com/pricing?utm_source=newsletter&utm_medium=email&utm_campaign=spring_promo&utm_content=header_cta
Everything from the ? onward is tracking data. The destination page works exactly as normal — the parameters are invisible to the visitor and have no effect on page behavior.
There are five standard parameters. Three are required by convention; two are optional but valuable once you’re tagging consistently.

| Parameter | Required? | Answers | Example values |
|---|---|---|---|
utm_source |
Yes | Who sent the traffic? | newsletter, facebook, google, partner_techblog |
utm_medium |
Yes | What channel type? | email, social, cpc, affiliate, referral |
utm_campaign |
Yes | Which initiative or promotion? | spring_promo, product_launch, weekly_digest |
utm_term |
No | What keyword or audience? | analytics+tools, remarketing_30d, audience_smb |
utm_content |
No | Which creative or link variant? | header_cta, footer_link, carousel_v2 |
utm_source identifies the specific platform, publication, or sender. Think of it as a proper noun: linkedin, google, newsletter, partner_techblog. It should name the thing, not describe the category — that’s what utm_medium is for.
utm_medium describes the marketing channel type. GA4 uses this value (along with source) to slot traffic into its default channel groupings. The values cpc, email, social, affiliate, and referral map directly to GA4’s built-in channel definitions. Deviate from those strings and your traffic may land in “Unassigned.”
utm_campaign ties the visit to a specific business initiative. It can be a promotion, a content series, a product launch, or a retention effort. This is the parameter you’ll filter on most often when asking “did this campaign work?”
utm_term was designed for paid search keywords but works for any targeting dimension: audience segments, interest categories, retargeting windows. Most teams only use it for paid search, where it captures which keyword triggered the ad.
utm_content is essential for A/B testing and placement tracking. If you include two links to the same page in one email — one in the header and one in the body — utm_content=header_cta vs utm_content=body_link tells you which one actually gets clicked.
Here’s the practical workflow I follow for every campaign, from link construction through to reporting.
Before touching any tool, write down the source, medium, and campaign name. Check your existing naming vocabulary — if your team uses linkedin for organic LinkedIn posts, don’t invent linkedin_organic for this campaign. Consistency across campaigns is more important than any individual naming choice.
Use Google’s Campaign URL Builder for one-off links. It’s free, validates required fields, and copies the final URL cleanly. Paste your destination URL, fill in the three required fields, add optional ones if relevant, then copy the result.

For a newsletter link to a pricing page, the builder output would look like:
https://example.com/pricing?utm_source=newsletter&utm_medium=email&utm_campaign=spring_promo
If you’re running paid social and want to track ad variants too:
https://example.com/pricing?utm_source=facebook&utm_medium=paid_social&utm_campaign=spring_promo&utm_content=carousel_v1&utm_term=website_visitors_30d
Paste the finished URL into a shared tracking spreadsheet before you use it anywhere. Record the destination, all parameter values, the campaign owner, and the date. When anomalous data appears in reports three months from now, this log is how you trace it.
Click it. Confirm the destination loads correctly. Open your GA4 DebugView (turn on debug mode via the GA4 debug configuration) and verify the session_source and session_medium parameters appear correctly. Catch typos before the campaign launches.
Send the email, publish the post, launch the ad. After 24 hours, check the Traffic Acquisition report in GA4 to confirm data is arriving under the expected source/medium combination.
This is where the work pays off. GA4 surfaces UTM data in several places, but two reports are the ones you’ll live in.
Go to Reports → Acquisition → Traffic Acquisition. By default this groups sessions by default channel grouping. Switch the primary dimension to “Session source / medium” or “Session campaign” to see your UTM values directly. This is your top-level view: which sources sent traffic, how many sessions, engagement rate, and conversions.
In most GA4 reports you can add a secondary dimension. Adding “Session source / medium” to a pages report, for example, shows you which campaigns sent traffic to each page. Adding it to a conversions report shows which UTM source drove the most goal completions.
One important note: GA4 distinguishes between session attribution (which campaign started the session) and event attribution (which campaign is credited for the conversion, based on your attribution model). For most teams doing campaign analysis, session source/medium is the right dimension. If you need to understand how AI-generated referral traffic gets attributed — a growing concern — UTMs become even more critical because referrer headers from AI tools are often stripped or missing.
UTM parameter values are case-sensitive strings. GA4 treats Email and email as two different mediums. Your Traffic Acquisition report will fragment into newsletter / Email, newsletter / email, newsletter / EMAIL, and Newsletter / email — all representing the same channel. Months of data, useless.
These four rules eliminate that problem:
utm_source=facebook not utm_source=Facebook.%20 and make URLs ugly and error-prone. Pick one separator and stick to it across your team.Your first-party data strategy only works if the inputs are clean. UTM naming consistency is the most basic form of data quality control.
This one corrupts data in a way that’s hard to detect later. If you put UTM parameters on links inside your own website — say, a blog post linking to your pricing page with utm_source=blog&utm_medium=internal — you override the original session source. A visitor who arrived from Google organic becomes attributed to “blog / internal” the moment they click that link. You’ve broken your attribution for that user’s entire session.
The rule is simple: UTMs are for external links only. Never tag links on your own domain.
One person writes utm_source=LinkedIn, another writes utm_source=linkedin, a third writes utm_source=linked_in. After six months you have four source variants for the same platform. The fix is documentation and enforcement — not just telling people, but using a shared builder with controlled inputs.
Using utm_campaign without utm_source and utm_medium creates partial data. GA4 may show the campaign name in one dimension but classify the session as “Unassigned” in channel groupings because it can’t determine the channel without medium. Always include all three required parameters together.
Long, descriptive names feel thorough when you’re building them. In reports, utm_campaign=2026_q1_lead_gen_ebook_download_enterprise_marketers_version_2_ab_test is impossible to scan. Keep campaign names short enough to read at a glance: ebook_enterprise_q1 is sufficient.
The most expensive mistake. Untagged links from newsletters, social posts, and partner sites arrive as “direct / none” — entirely invisible in channel-level analysis. If you’re spending budget on email or social, and those sessions show up as direct, you cannot demonstrate ROI. Tag everything external.
These are the tagging patterns I actually use, tested against GA4’s default channel groupings.
Newsletter / Email campaign:
?utm_source=newsletter&utm_medium=email&utm_campaign=weekly_digest&utm_content=header_cta
Organic LinkedIn post:
?utm_source=linkedin&utm_medium=social&utm_campaign=thought_leadership&utm_content=article_share
Facebook paid social ad:
?utm_source=facebook&utm_medium=paid_social&utm_campaign=retargeting_q2&utm_content=carousel_v2&utm_term=website_visitors_30d
Guest post / partner site:
?utm_source=partner_techblog&utm_medium=referral&utm_campaign=guest_post_march
QR code on printed material:
?utm_source=qr_code&utm_medium=offline&utm_campaign=conference_booth&utm_content=brochure
Notice how each example uses consistent lowercase, underscores instead of spaces, and medium values that map to GA4’s built-in channel groupings. That mapping matters: GA4 uses your utm_medium value to determine whether a session goes into “Email,” “Organic Social,” “Paid Social,” or “Referral” in the Traffic Acquisition report.
A common source of confusion is the difference between “Session source / medium” and “First user source / medium” in GA4.
Session source / medium — the source that started this specific session. If someone visited your site from Google last week (session 1) and returned via your newsletter today (session 2), session 2 is attributed to the newsletter, not Google.
First user source / medium — the source that acquired the user for the first time ever. The newsletter session above would still show Google as the first user source.
For campaign performance analysis, session source/medium is almost always the right dimension. You want to know which campaign drove sessions and conversions, not just which channel acquired each user originally. For broader acquisition strategy questions — which channels bring in new users who convert — first user source/medium is more relevant.
Your GA4 setup influences how cleanly this data appears. If you’re still working through the basics, the GA4 setup guide walks through the configuration steps that ensure UTM data is captured correctly from the start.
Before any campaign goes live, run through this:
Knowing how to use UTM parameters is one of the highest-leverage things you can do for campaign analytics. The technical barrier is low — the Campaign URL Builder does the construction for you. The real discipline is consistency: lowercase values, documented vocabulary, no internal tagging, and a log of every URL you send into the world.
Do that consistently for 60 days and you’ll have clean attribution data that actually answers the question “where did my traffic come from?” GA4’s Traffic Acquisition report will show you source/medium combinations that mean something, campaign names you recognize, and content variants you can compare. That’s the difference between a dashboard you ignore and one you make decisions from.
Start with the next email or social post you send. Tag it properly. Build the habit before you try to build the system. The Campaign URL Builder is one tab away. The GA4 platform records every tagged click automatically. There’s no infrastructure to set up. You just have to start.
First published . Last updated .