Medium in marketing analytics is the dimension that describes how a visitor arrived β the mechanism of the click, not the specific origin. In Google Analytics 4, medium pairs with source to form the source/medium combination that powers every channel-grouping decision GA4 makes. This guide covers the UTM medium parameter and its standard values, the source/medium/campaign distinction, GA4’s default channel-mapping rules, naming conventions that won’t break attribution, common tagging mistakes (case sensitivity, typos, blanks), and how medium flows through multi-channel funnels β plus a FAQ answering the questions analysts actually get asked about utm_medium.
What Is Medium in Marketing Analytics
Medium is a campaign-tracking dimension that describes the general category of traffic β organic search, paid clicks, email, social, referral, affiliate, display. While source identifies the specific origin (google, facebook, mailchimp), medium describes the mechanism of the visit. The two together form the source/medium pair that GA4 uses to bucket every session into a channel.
Two examples make the distinction concrete:
- Source
google+ mediumorganicβ unpaid search click β channel Organic Search. - Source
google+ mediumcpcβ paid Ads click β channel Paid Search.
Same source, different medium, different channel. Get medium wrong and the entire attribution chain misclassifies β you might see a paid-search ROI report counting organic visits, or organic-search performance polluted by tagged email links.
UTM Medium Parameter Explained
The utm_medium parameter is one of five UTM tags Google supports for tracking marketing links: utm_source, utm_medium, utm_campaign, utm_content, and utm_term. You append it to a destination URL as a query string, and GA4 reads it on arrival to assign session-level dimensions.
A well-tagged URL looks like this:
https://example.com/landing?utm_source=mailchimp&utm_medium=email&utm_campaign=2026-q2-launch
When the visitor lands, GA4 records session_source = mailchimp, session_medium = email, and session_campaign = 2026-q2-launch. These dimensions persist for the duration of the session and flow into every report β Acquisition, Explorations, BigQuery exports, conversion paths.
The convention you choose for utm_medium directly determines whether GA4’s default channel grouping recognises the visit. Stick to Google’s documented values (organic, cpc, email, social, referral, affiliate, display) and channels resolve cleanly. Invent your own (utm_medium=newsletter) and the visit lands in (unassigned). Use the official Google Campaign URL Builder to tag links with the correct schema.
Common Medium Values and What Each One Means
The standard utm_medium taxonomy covers seven core categories. Anything outside this list either gets auto-detected by GA4 (organic, referral, direct) or risks falling into Unassigned.
| utm_medium | Traffic type | GA4 default channel | Typical use case |
|---|---|---|---|
organic |
Unpaid search engine traffic | Organic Search | Auto-detected from Referer; rarely tagged manually |
cpc |
Paid search clicks (cost-per-click) | Paid Search | Google Ads auto-tagging via gclid; manual tags for Bing/Yahoo |
email |
Newsletter, drip, transactional emails | All Mailchimp, Klaviyo, SendGrid, ConvertKit links | |
social |
Organic social media posts | Organic Social | Tagged links in tweets, LinkedIn posts, Reddit comments |
paid-social |
Paid social advertising | Paid Social | Facebook Ads, LinkedIn Ads, X Ads campaigns |
referral |
Inbound links from other websites | Referral | Auto-detected from Referer header; tag if you want explicit attribution |
affiliate |
Affiliate partner clicks | Affiliates | Impact, PartnerStack, Refersion, custom affiliate programs |
display |
Banner / programmatic ads | Display | Google Display Network, programmatic exchanges |
video |
Video platform traffic | Video / Organic Video | YouTube card links, Vimeo embeds, sponsored video posts |
(none) |
No Referer, no UTM | Direct | Typed URL, bookmark, dark social, app deep link |
Medium vs Source vs Campaign β Three Dimensions, One Job
Source, medium, and campaign answer three different questions about each visit. Confusing them is the most common cause of broken attribution reports. Here’s the clean reference:
| Dimension | Question it answers | Granularity | Example |
|---|---|---|---|
| Source | Where exactly did the click come from? | Specific origin (vendor, hostname, publication) | mailchimp, facebook, nytimes, github |
| Medium | What kind of channel is this? | Broad category (mechanism of click) | email, cpc, referral, social |
| Campaign | Which specific marketing initiative? | Time-bounded effort (launch, promo, A/B variant) | 2026-q2-launch, black-friday, welcome-series-v3 |
The hierarchy: medium is the channel category, source is the platform, campaign is the specific activation. A Mailchimp newsletter promoting a Q2 product launch tags as utm_source=mailchimp&utm_medium=email&utm_campaign=q2-launch. Each dimension answers its own analytical question, and reports stay clean because each level has clear semantics.
Practical rule: if you’re tempted to put product or seasonal info in medium (“utm_medium=blackfriday”), stop β that belongs in utm_campaign. Medium should describe how, not what.
How GA4 Default Channel Group Maps Medium to Channels
GA4 evaluates each session against an ordered ruleset to pick a channel. The medium value drives most of the decisions. The simplified flow:
- Cross-network if Performance Max or Smart Shopping campaign is detected (gclid + special signals).
- Paid Search if
utm_mediummatchescpc|ppc|paidsearchregex, or gclid is present. - Paid Social if medium matches
paid-social|paidsocialregex. - Paid Video if medium matches
cpv|cpm|paidvideo. - Display if medium matches
display|banner|cpm. - Email if medium matches
email|e-mail|e_mail|e mail. - Affiliates if medium =
affiliateexactly. - Organic Search / Social / Video if Referer matches a known list (and no paid signal).
- Referral if external Referer survives all earlier rules.
- Direct if no Referer and no UTM.
- Unassigned if nothing matches.
A custom value like utm_medium=newsletter matches no rule above and lands in Unassigned. To fix without renaming, build a custom channel group under Admin β Data display β Channel groups and add a rule: “if medium contains ‘newsletter’ β Email”. For most teams it’s faster to just standardise on email.
Standard Medium Naming Conventions
A defensible utm_medium taxonomy has four properties: lowercase only, hyphen-separated multi-word values, GA4-recognised core values, and documented exceptions for any custom additions.
- Always lowercase. GA4 is case-sensitive.
Email,EMAIL, andemailare three separate dimension values that fragment your reports. - Hyphens, not underscores or spaces.
paid-socialis the convention. Avoidpaid_social(varies across vendors) andpaid social(URL-encoded becomespaid%20social). - Pick from the standard list.
organic,cpc,email,social,paid-social,referral,affiliate,display,video. Anything else needs a custom channel group rule. - Document exceptions in writing. If marketing wants
utm_medium=podcast, add it to a shared doc, define which channel it should map to, and create the matching custom rule on day one.
For technical guidance, Google’s channel grouping documentation is the canonical reference. Cross-check against the Search Central docs when sorting out the boundary between organic search and self-attributed traffic.
Tracking Medium in GA4 Reports
GA4 surfaces medium across multiple report surfaces. Three are essential for day-to-day acquisition analysis:
- Reports β Acquisition β Traffic acquisition. Default view groups by Session default channel group. Switch the primary dimension to Session medium to see raw counts per medium value across the property. Useful for spotting case-sensitivity bugs (Email vs email vs EMAIL appearing as separate rows).
- Reports β Acquisition β User acquisition. Same shape but attributed to first-touch instead of last-touch β answers “which medium discovered each new user” vs “which medium triggered this session”.
- Explore β Free Form. Add Session source / medium as the row dimension and conversions, events, and revenue as metrics. This is the workhorse for medium-level performance analysis. Filter Session medium =
emailfor newsletter ROI; =cpcfor paid-search efficiency.
For programmatic analysis, the BigQuery export exposes traffic_source.medium, session_traffic_source_last_click, and collected_traffic_source.manual_medium as separate fields. Querying these in BigQuery or Looker Studio bypasses the cardinality limits of the GA4 UI and lets you build custom medium reports across millions of sessions.
Common Medium Tagging Mistakes
Eight of every ten medium-related attribution bugs come from the same handful of mistakes. Audit any new tagging convention against this list before launching.
- Inconsistent case.
Email,email, andEMAILcreate three rows in every report. Lowercase everything, always. - Typos.
emial,cpv(when you meantcpc),refferal. These end up in Unassigned. Use the URL Builder, not free-form typing. - Blank or missing utm_medium. If
utm_sourceis set bututm_mediumis empty, GA4 cannot map the visit to a channel. Both must be present together β or neither. - URL-encoded spaces.
utm_medium=paid%20socialrenders aspaid socialin reports and confuses channel rules. Use hyphens. - Custom values without channel rules.
utm_medium=newsletter,=partner,=podcastall land in Unassigned unless you add custom channel-group rules. - Tagging internal links. Adding UTM tags to navigation links inside your own site creates self-referrals and breaks session continuity. Never tag internal links.
- Mixed case from copy-paste. Marketing copies a tagged URL into a doc; the doc’s auto-formatter capitalises a letter; the link goes out in a campaign with
utm_medium=Emailinstead ofemail. Add a linter or pre-flight check. - Putting campaign info in medium.
utm_medium=blackfridayconflates two dimensions. Black Friday is a campaign, not a medium. Useutm_medium=email&utm_campaign=blackfriday.
Medium in Multi-Channel Funnels and Attribution
Medium is one of three identifiers GA4 uses to build conversion paths. When a user touches your site multiple times before converting β say, organic search β email β direct β paid search β purchase β each touchpoint records its own session_medium, and GA4’s attribution model decides how to distribute credit.
The data-driven attribution model (default in GA4) uses Shapley values to weight each medium proportionally to its contribution. Email might get 30% credit, organic 25%, paid 30%, direct 15% β depending on the historical pattern of paths that converted. Switch to last-click in Advertising β Attribution settings to see only the final touchpoint, or first-click for discovery attribution.
Two practical implications:
- Medium quality matters more than medium volume. A medium with 5% of sessions but 15% of attributed revenue is more valuable than one with 30% of sessions and 10% of revenue. Always pair Sessions with Conversions and Total revenue when ranking mediums.
- Reports β Advertising β Conversion paths is the canonical view for understanding how mediums chain. It shows the most common N-touch sequences and which medium combinations convert. Useful for budget allocation across channels rather than evaluating each in isolation.
For deeper analysis at the user level, combine medium with cohort analysis to see how first-touch medium predicts long-term retention and revenue. A user acquired via email often has different LTV than one acquired via paid social, even at the same conversion event.
Best Practices for Medium Taxonomy
A medium taxonomy that survives a year of campaigns has these guardrails:
- Document the canonical list. Maintain a one-page reference (Notion, internal wiki, README in your tagging repo) that lists every approved utm_medium value with its meaning, example URL, and the GA4 channel it maps to.
- Use a URL Builder, not free-form typing. Whether it’s the official Campaign URL Builder, an internal tool, or a spreadsheet template, force every campaign URL through a controlled vocabulary.
- Audit weekly with the unassigned report. Filter Reports β Acquisition by Session default channel group = Unassigned. Each row is a tagging gap. Investigate, fix the source, add custom channel rules if needed.
- Treat medium as immutable. Once a medium value is in production, never rename it β historical reports break. If you need to retire a value, replace it forward and keep the old one alive in your channel-group rules.
- Pair medium with content for variant tracking. Use
utm_content(not utm_medium) to differentiate ad variants, button positions, or A/B test arms within the same medium. Medium stays clean. - Standardise across vendors. Mailchimp, HubSpot, Klaviyo, Marketo all default to
utm_medium=emailβ verify this in each tool’s auto-tagging settings. Don’t let a vendor decide your taxonomy for you.
The payoff for taxonomy discipline is honest reporting. When CMO asks “what’s email contributing this quarter?”, the answer is one filter (session_medium = email) instead of a regex query trying to catch email|Email|EMAIL|newsletter|drip. Build the discipline early and you get years of trustworthy comparison.
Frequently Asked Questions
What is utm_medium in Google Analytics?
utm_medium is a URL parameter that tells GA4 the general category of traffic β organic, cpc, email, social, referral, affiliate, display. GA4 reads it on arrival and stores it as session_medium, then uses it to map the visit to one of the default channel groups (Email, Paid Search, Organic Search, etc.) for reporting.
What are the standard utm_medium values?
The seven core values GA4 recognises by default are: organic, cpc, email, social, paid-social, referral, affiliate, and display. Two more (video and (none)) cover video platforms and direct traffic. Anything outside this list needs a custom channel-group rule or it lands in Unassigned.
What is the difference between source and medium?
Source identifies the specific origin (mailchimp, facebook, nytimes, github) β the vendor or publication. Medium describes the mechanism category (email, social, referral). Both must be present together for a visit to map cleanly to a channel. Source is granular; medium is broad.
Why is my utm_medium showing as unassigned?
“Unassigned” means GA4 received a medium value that doesn’t match any default channel-group rule. Common causes: typos (emial), inconsistent case (Email), or custom values like utm_medium=newsletter or utm_medium=partner. Fix by either standardising to a recognised value or adding a custom channel-group rule under Admin β Data display β Channel groups.
Is utm_medium case-sensitive?
Yes. Email, EMAIL, and email are three distinct values in GA4 reports β they fragment your data and break channel mapping (only lowercase email matches the default rule). Always use lowercase for utm_medium and document the convention so contractors and tool defaults don’t drift.
Can I use a custom utm_medium value?
You can, but you’ll need a matching custom channel group rule, otherwise the traffic lands in Unassigned. Go to Admin β Data display β Channel groups and create a rule like “if Session medium contains ‘newsletter’ then channel = Email”. For most teams the simpler fix is to standardise on the GA4-recognised values (email, social, cpc, referral) and use utm_campaign for the variant-level information.
How does utm_medium affect attribution?
Medium is one of three dimensions GA4’s attribution model uses to build conversion paths. Each touchpoint in a path records its own session_medium; the model (data-driven by default, or last-click / first-click) distributes conversion credit across mediums based on their contribution. Cleaner medium tagging means more accurate path analysis and budget allocation across channels.
Related Terms
- UTM parameters β the five tags (source, medium, campaign, content, term) GA4 reads from URLs
- Campaign β utm_campaign grouping for time-bounded marketing initiatives
- Referral traffic β visits classified with medium = referral
- Attribution β how GA4 distributes conversion credit across mediums and touchpoints
- Attribution model β last-click vs first-click vs data-driven, applied across mediums
- Events β the user actions tracked alongside medium for conversion analysis
- Conversions β events used to measure the value each medium contributes
- Macro conversions β high-value goals where medium quality matters most
- Cohort analysis β combine first-touch medium with retention curves for LTV insights
- BigQuery export β raw
traffic_source.mediumfield for custom medium reporting - Looker Studio β build custom medium dashboards beyond the GA4 UI