Here's an uncomfortable pattern. Someone spends an hour a week submitting their posts to bookmarking sites and communities, keeps doing it for six months, and has no idea whether a single visitor ever came from it. The activity feels productive. Nobody ever checks.
Measurement is what turns sharing from a ritual into a decision. And it's mostly a one-time setup: tag your links so they identify themselves, then read two reports. Once that's running, the question "is this platform worth it?" has an actual answer instead of a vibe.
What Your Analytics Can and Can't See
When a visitor arrives, the browser may send a referrer — the page they came from. Your analytics groups visits by that value: search engines become "organic", other websites become "referral", and visits with no referrer at all become "direct".
Three things regularly destroy that signal:
- The click didn't come from a web page. Links opened from a desktop email client, a messaging app, or a PDF often arrive with no referrer.
- The source page told the browser not to send one. Many platforms apply a referrer policy or route outbound clicks through a redirect, which can strip or replace the original source.
- The protocol changed. A link on an insecure page pointing to a secure one may lose the referrer entirely.
This is why "direct" traffic is usually far larger than the number of people typing your URL by hand. It's the bucket where unattributed traffic goes to die — the phenomenon people call dark social. You cannot fix it by hoping. You fix it by tagging.
Tag Every Link You Share
UTM parameters are labels you bolt onto the end of a URL. Your analytics reads them and files the visit exactly where you told it to, whether or not a referrer survived.
https://example.com/your-post?utm_source=pinboard&utm_medium=bookmark&utm_campaign=spring-guide
Three of the five parameters do almost all the work:
- utm_source — the specific place: the platform, newsletter, or community name.
- utm_medium — the kind of place:
bookmark,social,email,forum,referral. This is what lets you compare channels rather than individual sites. - utm_campaign — the thing you're promoting, so one push can be measured across every place you shared it.
utm_content and utm_term are optional; use utm_content to distinguish two variants of the same link in the same place.
Consistency is the whole game. utm_source=Reddit, utm_source=reddit, and utm_source=reddit.com are three different sources in your reports, and your data will be a mess within a month. Pick lowercase, no spaces, and write your conventions down — a five-row note listing your standard sources and mediums is enough. Build the URLs with a tagging tool rather than by hand; typos are silent and permanent.
Two rules that prevent self-inflicted damage:
- Never put UTM parameters on internal links between pages on your own site. Doing so restarts the session attribution and makes your own site look like its own traffic source.
- Never tag a link you're submitting as a canonical reference where the parameterised URL might get indexed or copied around. For sharing links that's rarely an issue; for anything that behaves like a permanent citation, share the clean URL and measure via referrer instead.
The Two Reports to Read
1. Traffic acquisition, grouped by source and medium. This tells you which places sent sessions. Look at it over a period long enough to matter — a single day of a single share tells you nothing.
2. Landing pages, filtered to that source. This tells you what they came for. A platform that sends people to one old post and nothing else is a very different asset from one that sends people to whatever you publish.
Then add the column that changes decisions: engagement, in whatever form your analytics gives you — time on page, scroll, pages per session, or a conversion event you've defined. Sharing that produces a hundred visitors who bounce in three seconds is worth less than five visitors who read to the end and subscribe. Judge platforms on the second number, always.
Read the Numbers Honestly
Four traps swallow most amateur analysis:
Bots and previews inflate everything. Link previews, uptime checkers, and crawlers can register as visits. If a platform "sent" you forty visitors with zero seconds on page, you found automated traffic, not an audience.
A spike is not a channel. One post doing well somewhere tells you that post did well. Two or three shares over a few weeks tell you whether the place works.
Self-visits pollute the data. Check your own link after posting, from your own device, every time — that's you in the report. Filter your own traffic if your analytics allows it.
Attribution isn't credit. Someone may discover you on one platform, look you up two days later, and arrive as "direct". The tagged visit is a floor on a platform's contribution, not a ceiling. Treat sharing measurement as a comparison between places, not as a full accounting of influence.
From Measurement to a Decision
Once you have a few weeks of clean data, the review is quick. For each place you share:
| Signal | What it suggests |
|---|---|
| Steady visits, decent engagement | Keep. Increase effort here. |
| High visits, near-zero engagement | Wrong audience — or bots. Check before celebrating. |
| Almost no visits, but real engagement from the few | Niche but real. Keep if it's cheap in time. |
| No visits at all after several genuine attempts | Stop. That hour goes somewhere else. |
The fourth row is the one people avoid, and it's the whole point of measuring. Most sharing routines contain at least one habit that has never worked and never will, kept alive purely because nobody looked.
Worth saying plainly: measurement will not turn a spammy approach into a working one. Blasting the same link into fifty places produces removals and bans, not traffic — the etiquette guide covers what earns a welcome instead. And if your goal is ranking rather than visits, be clear-eyed about what these links do and don't do for search; that's a separate question with a separate answer.
A 20-Minute Setup
- Confirm your analytics is installed and recording on every page.
- Write your naming convention: a list of
utm_sourcevalues you'll use, and a small fixed set ofutm_mediumvalues. - Build tagged URLs with a link builder, not by hand.
- Optionally, put tagged links behind a short-link tool so they look clean when shared — just make sure the shortener passes the parameters through to the destination.
- Diary a review for four weeks out. Without the diary entry, this doesn't happen.
That's it. The hard part was never the tooling; it's committing in advance to act on what you find.
Frequently Asked Questions
What are UTM parameters? Labels appended to a URL — source, medium, campaign — that your analytics reads to attribute the visit. They work even when the browser sends no referrer.
Why is so much of my traffic labelled "direct"? Because the referrer was missing: clicks from apps, email clients, documents, or platforms with a strict referrer policy. Tagging your own shared links moves them out of that bucket.
Do UTM parameters hurt SEO? They create URL variants, which is why you shouldn't use them on internal links or anywhere the tagged URL could become the version people cite. For outbound sharing links, they're standard practice.
Should I use a link shortener? It's fine, and useful for tidy display and click counts, as long as it preserves your parameters and the redirect is permanent enough not to break later. Keep a record of what each short link points to.
How long before I can judge a platform? Several genuine shares over at least a few weeks. One post is an anecdote; a pattern across posts is a signal.
Measure First, Then Commit
You don't need a dashboard or a data team — you need tagged links, two reports, and a calendar reminder to look. That's the difference between sharing as a habit and sharing as a strategy.
If you're comparing analytics, tagging, or short-link tools to set this up, save the shortlist into a list on BookmarkDiscover so the options are side by side when you decide.
