A UTM naming convention your whole team will follow

Open the UTM tab of your Sources report. If you see something like this:

(Sample data)

you don't have an analytics problem, you have a naming problem. Those four rows are one channel, and only their sum is the real figure. Worse, it happens quietly until someone at the quarterly review asks why Facebook performed so badly.

UTM values are case-sensitive and space-sensitive, and analytics can only group by the characters it receives. The only fix is for everyone to use the same rules before the link is built.

What each of the five parameters is for

First, agree on what the parameters mean; a lot of mess comes from mixing them up:

• utm_source — the platform or site the traffic comes from: google, newsletter, partner_acme.
• utm_medium — the type of traffic: cpc (pay per click), email, social, referral.
• utm_campaign — which promotion it belongs to: 2026_q4_black_friday.
• utm_term — the paid search keyword, or an audience you want to distinguish.
• utm_content — which creative or placement within the campaign: banner_top, video_v2.

A way to remember it: source is who, medium is how, campaign is why, content is which one.

Five naming rules

1. Everything lowercase

Facebook and facebook are two rows in a report. Every value is lowercase, no exceptions.

2. Underscores, not spaces

Spaces turn into %20 or + in URLs, and tools treat them differently. Join words with underscores — black_friday, never a mix of black friday and black-friday. Pick one separator and stick with it.

3. Sources are platform names, not domains

Write google, not google.com or www.google.com. Write linkedin, not lnkd. Give each common platform one spelling and write it down in the sheet below.

4. Medium comes from a short fixed list

Medium is the parameter that gets out of hand fastest. We suggest allowing only these:

Anything else isn't allowed. Your channel report stays very clean.

5. Campaign names include the date

spring_sale will be used again next year and the two campaigns will blur together. Use 2026_03_spring_sale — year and month first, so sorting by name also sorts by date.

A sheet you can use as is

Rules in a document are soon forgotten. What works better is a sheet everyone can see, with a row for every new link:

(Example)

Three benefits: newcomers copy existing rows instead of inventing spellings; at review time you can see who built each link, when and why; and a mistake can be traced straight away.

To build links, use the UTM link builder; it lowercases values and flags common mistakes.

Common traps

Don't tag internal links. If a banner on your home page links to your own campaign page with UTM tags, the visitor's session is split and the source rewritten to whatever you typed — the credit for the search visit that brought them is lost. UTM tags belong on external links pointing to your site.

Paid ads don't always need manual tags. Major ad platforms append a click ID to the landing URL, and analytics uses it to put the visit in the paid channel even without UTM tags. Manual tags are worth adding to tell campaigns and creatives apart, not to mark traffic as paid.

Check that shortened links keep their parameters. Some shorteners and redirects drop them. Click the finished link yourself and check the source in Real-time.

Already a mess?

If your reports already hold a pile of inconsistent history, you don't have to merge it by hand. Add channel rules that map fb, Facebook and facebook.com to one source, then run a recalculation of history. You write the rule once, and future misspellings are tidied automatically too.

Rules are a safety net, though, not a substitute. The real fix is that from the next link onwards, everyone works from the same sheet.

Related articles

Marketing attribution·Sep 25, 2026·4 min read

First, last, linear, position, time decay: what each attribution model answers

No attribution model is right or wrong — only suited or not to the question you're asking. One set of customer journeys, five models, five different answers about who gets the credit.

Read article

Product notes·Oct 16, 2026·3 min read

Why bot traffic gets its own report

By default we take identified bots out of your human reports rather than dropping them or mixing them in. The three reasons behind that, and what it means for your numbers.

Read article

Playbooks·Oct 14, 2026·5 min read

Checkout drop-off: a seven-step investigation from data to fix

Plenty of add-to-carts, not many payments. A checklist to work through in order — rule out technical problems first, then find which step, which visitors and what cause.

Read article