Where Your Visitors Came From

The four buckets

Organic search. Somebody searched and clicked a result that was not an advertisement.

Referral. Somebody clicked a link on another website.

Social. A link from a social platform, which is really a subset of referral.

Direct. Everything else.

That last one is the problem, because it sounds like a category and is actually a residue.

What is really in direct

• Typing the address or using a bookmark, which is the intended meaning.
• Clicking a link in an email, which frequently arrives untagged.
• Clicking a link in a messaging app, which usually passes nothing.
• A link in a PDF or a document.
• Anywhere the referring information was stripped, which happens routinely.
• QR codes and printed addresses, which carry no source at all.

Which means a large direct figure is not evidence that people know your brand. It is mostly evidence of traffic whose origin was not recorded.

A worked example

A business whose report showed almost sixty percent direct traffic, which they interpreted as strong brand recognition.

Checking the actual sources told a different story.

They had recently put their address on van signage, on invoices, and in an email signature, and had run a print advertisement.

All of that arrives as direct, because a person typing an address carries no origin.

Adding a short trackable address to the print advertisement, and tagging the links in their email footer, moved a meaningful chunk of that traffic into identifiable categories within two months.

The brand recognition interpretation had been comforting and wrong, and it had nearly prevented them from repeating a print advertisement that was genuinely working.

Tagging your own links

The fix for most of the direct problem, and it costs nothing.

Adding parameters to a link tells analytics where it came from, and the link still works normally.

Worth tagging: links in your email signature, links in any newsletter, links from your own social profiles, links in a supplier's directory listing, and anything in a document you send out.

Keep the naming consistent, because inconsistent tags produce a report listing the same source five ways.

And do not tag internal links between your own pages, which resets the session and corrupts the original source, a common and unhelpful mistake.

Offline sources

The category no tagging can reach, and which for many trades is most of the business.

Van signage, a board on a site, a printed advertisement, a card in a shop, word of mouth.

The practical solution is a short memorable address that redirects to the relevant page, used only in that one place.

A single word after your domain is enough, and traffic arriving that way is unambiguous.

The alternative, and the more reliable one, is asking every enquiry how they heard of you and writing it down, which captures what no measurement can.

Reading referral traffic

Usually the most useful bucket for a small business and the least looked at.

It tells you which directories, associations and partners are actually sending people, as opposed to which ones you are listed in.

Frequently the answer is that most listings send nothing and one sends a steady trickle, which is worth knowing before renewing any of them.

It also reveals sites linking to you that you did not know about, which is occasionally an opportunity and occasionally something to look at more carefully.

Worth reviewing twice a year rather than monthly, since the list changes slowly.

Reading organic search traffic

The bucket most businesses care about and the one most affected by things outside the site.

A rise or fall here can mean your pages changed, or that a competitor's did, or that the searches themselves changed in volume.

Seasonal trades see swings that have nothing to do with anything they did, and reading a summer drop as a failure produces unnecessary work.

Which is why organic traffic is worth comparing against the same month last year rather than against last month.

And worth splitting between searches for the business name, which are people who already knew you, and everything else, which is genuine discovery.

Those two behave differently and a report combining them can hide a fall in one behind a rise in the other.

The counter-case

Where source data matters less.

A business whose enquiries arrive almost entirely by phone from referral, where the website is confirmation rather than acquisition.

Anywhere the volume is too small for the split to be meaningful. Sixty sessions a month divided four ways produces numbers that move for no reason.

And any situation where the honest answer is already known, because the owner speaks to every customer and knows how each found them.

In that last case the analytics is confirming something better evidence already established, which is fine and not worth much effort.

Search terms, and what happened to them

A gap worth knowing about, because people look for this and cannot find it.

Analytics will tell you that somebody arrived from a search. It will not tell you what they searched for.

Those terms were withheld from analytics years ago for privacy reasons, and what remains in most reports is a large entry reading not provided.

The terms are available, in aggregate, in Search Console, which is a separate tool reporting on search specifically.

Which means the answer to what did people search for lives in one place and the answer to what did they do next lives in another, and connecting them takes a deliberate step.

For a small business, reading the Search Console query list once a quarter is sufficient and frequently more useful than anything in the analytics report.

What the report cannot tell you

Worth stating plainly.

It cannot tell you which source produced revenue, only which produced visits.

A source sending a hundred sessions that never enquire is worth less than one sending five that do, and the report shows the first as ten times better.

Connecting source to money requires either tracking enquiries as events, which is a further step, or asking people, which is simpler and more reliable.

For most small businesses, asking is the honest answer and the report is supporting evidence.

What to do with this

1. Check what proportion is direct.
2. Tag your email signature and newsletter links.
3. Set up a short address for anything printed.
4. Review referrals twice a year.
5. Ask every enquiry how they found you.
6. Compare the two, which is where the surprises are.

The sixth frequently shows that a channel producing little measurable traffic is producing a disproportionate share of actual work.

How to count the outcome rather than the visit is covered in measuring whether email is working.

Frequently asked questions

West Coast Media Solutions Inc. provides web design, web development, hosting, digital marketing, and business consulting to organisations across Canada, drawing on more than twenty-five years in the field.