Reading click attribution: source, medium, and intent

The three layers

Every click has three pieces of attribution data, and they tell you different things.

• Source — where did the click happen? (newsletter, instagram, qr-print, etc.)
• Medium — what kind of channel was it? (email, social, paid, organic)
• Intent — what was the visitor trying to do? (browse, buy, sign up, contact)

Source and medium are usually set via UTM parameters. Intent is inferred from behavior post-click: bounce rate, scroll depth, conversion path.

UTMs done right

Three habits to adopt now:

• Use lowercase, hyphenated values (utm_campaign=spring-launch, not "Spring Launch")
• Pick a finite vocabulary for source and medium. Document it. Enforce it via presets.
• Never put email addresses or session IDs in UTMs — they leak into analytics

Referrer headers are an imperfect supplement

HTTP referrers give you context UTMs can miss — but they also drop silently when traffic comes from HTTPS to HTTP, from apps with in-app browsers, or from privacy-respecting browsers. Treat the referrer as additional context, not the source of truth.

"Stop fighting over last-click vs first-click. Just instrument the journey and read the patterns. The model debate matters less than getting clean data in."

Reading the patterns

Once data is clean, three patterns are worth watching:

• Time-to-conversion by source — paid social often clicks fast but converts late.
• Assisted conversions — newsletter rarely closes deals alone, but it warms the path.
• Device shifts — mobile click, desktop convert is now the default journey for B2B.

Where OpenMyLink fits

OpenMyLink captures click-side attribution at the redirect: source, medium, campaign, referrer, country, device, and any custom parameters. Fire tracking pixels (Meta, Google Ads, LinkedIn, TikTok, Reddit, Snap, Pinterest) on every click so the conversion side picks up the trail, and pull per-link or per-campaign rollups via the REST API at openmy.link/api.