UTM builder

A UTM builder should leave you with more than one link

A campaign URL is useful for the minute it is created. The naming decision behind it needs to survive for weeks: which source label the team chose, whether email was called email or newsletter, and how the campaign name was separated. That is why this builder returns three connected artifacts. The URL is ready to publish, the JSON template preserves the campaign fields for the next destination, and the CSV is a reviewable campaign row for a spreadsheet or handoff.

The template deliberately leaves out the destination URL. Reusing a campaign should keep its naming convention, not quietly keep last week's landing page. Paste or open the template here, add the new destination, and inspect the complete generated address before copying it. Nothing is saved in an account, so the downloaded template is the durable copy.

How the campaign URL is assembled

The destination is parsed as a real web URL rather than joined with a question mark by hand. Existing non-campaign parameters stay in place, as does a fragment such as #pricing. Existing parameter names beginning with utm_ are removed case-insensitively before the visible fields are added. That matters when a landing-page link has been copied from an earlier campaign: leaving both old and new source parameters would make the reporting result depend on which duplicate a later system happens to read.

Each field is encoded as a query value. An ampersand inside a source is data, not the start of another parameter; a plus sign inside a medium stays a literal plus; spaces, slashes, Unicode and hash characters cannot escape into the rest of the URL. The displayed link is the exact string written into the CSV. There is no hidden shortener and no redirect layer. If you need to inspect a link someone else built, URL decode explains the existing query string instead of changing it.

Source, medium and campaign are naming decisions

Source answers where the visit originated: a newsletter name, partner, platform or publication. Medium describes the channel class, such as email, paid social or referral. Campaign groups the work you want to compare. The builder requires those three because a blank value creates an incomplete reporting label, but it does not force lowercase or replace your spaces. A tool cannot know whether your existing convention uses fall-launch, fall_launch or Fall Launch. Silently choosing one would create a second convention rather than enforce yours.

Campaign ID, term and content are optional. ID can join the link to an external campaign record. Term is often used for a paid-search keyword, while content distinguishes links that share the same destination and campaign, such as a hero button and a footer link. Empty optional parameters are omitted rather than emitted as meaningless blank keys.

What the verified CSV does and does not prove

Spreadsheet programs may interpret cells beginning with =, +, - or @ as formulas. Campaign labels are text, so those prefixes are neutralized in the CSV before download. The file uses a UTF-8 byte order mark and CRLF rows for broad spreadsheet compatibility. It is then parsed again in this tab, and all eight headers and every cell are compared with the intended campaign row. The green receipt means the generated table survived that exact local round trip.

It does not mean a spreadsheet imported the file, that a link was clicked, or that an analytics property recorded a session. This page never fetches the destination and never contacts Google Analytics. After publishing, use GA4 DebugView or your own real-time checks to confirm measurement, and use the Search Console report analyzer only for the separate question of how organic queries and pages performed. Those are evidence from live systems; a URL builder cannot manufacture it.

Keep campaign structure separate from campaign performance

A well-formed URL can still point to the wrong page, carry a naming convention nobody else follows, or be published where a redirect strips its query string. Review the visible URL, open it in the context where it will actually appear, and compare the resulting data after the campaign starts. If rankings or impressions are the concern, the Google ranking diagnostic walks through that evidence without treating UTM parameters as a ranking fix.

The same boundary applies to structured publishing data. A FAQ schema generator can produce valid JSON-LD while still being unable to promise a search presentation. Likewise, this builder can prove its encoding and exports while making no promise about analytics delivery. When page speed is the suspected issue instead, the Lighthouse report analyzer reads an actual report rather than inferring performance from the campaign URL.