Count what already happens: form submissions, phone calls, quotes issued, jobs won. Add server logs and search console for context. None of it requires tracking an individual.

The question analytics is being asked to answer

Almost always the same one: is this working, and where is the work coming from.

That question does not require following anybody. It requires counting outcomes and knowing roughly what preceded them.

Most small businesses already have the outcome data, sitting in systems nobody thinks of as measurement, and they go looking for it in a visits report instead.

Start at the end, not the beginning

The most useful reframing available in small business measurement.

Visits are the top of the process and the least reliable thing to count. Enquiries, quotes and jobs are the bottom of it, and they are recorded properly because money depends on them.

Nobody loses an invoice. Plenty of businesses lose a month of analytics and never notice.

So build your measurement from the reliable end upward: how many jobs, how many quotes, how many enquiries, and only then how many visits.

What you already hold

Every one of these is first-hand, complete, and unaffected by consent banners, ad blockers or browser changes.

Server logs

Your web server records every request it serves, and that record exists whether or not anything is added to the page.

It tells you which pages were requested, how often, what referred them, and what errors were returned.

Nothing can block it, because it is not running in the visitor's browser.

The trade is noise. A large share of raw log traffic is automated, from crawlers, scanners and monitoring, and reading logs without filtering that out produces numbers that are wildly wrong in a flattering direction.

Most hosting control panels include a basic log report, which is enough for a small site, and it is worth looking at once if only to see how much of your traffic was never a person.

The search console

Free, aggregated, and it tracks nobody individually.

It reports which searches showed your pages, how often, how often they were clicked, and which page was shown.

That is the single most useful non-tracking source available, because it describes intent rather than behaviour: what people were actually looking for when they found you.

It also covers the visitors your analytics missed, since it is measuring the search results rather than your page.

Counting by hand is legitimate

Worth saying because there is a persistent belief that a number is only real if a system produced it.

A tally of enquiries kept in a spreadsheet, updated weekly, is more accurate than most automated conversion tracking and takes about four minutes.

For a business receiving thirty enquiries a month, hand counting is entirely practical and has the enormous advantage that the person counting notices things: an unusual cluster, a repeated question, a source nobody expected.

Automation is worth having when volume makes counting impractical. Below that, it mostly adds a layer between you and the thing you are trying to understand.

A worked example

A two-van trade business had analytics they never opened and no idea whether their site produced work.

They started a single sheet: date, name, how they found us, quoted, won, value.

Every enquiry went on it, from any source, filled in by whoever took the call.

After five months they had ninety-one rows, and the picture was unambiguous. Just over half came from the business listing, a quarter from repeat customers and referrals, and about a fifth from the website.

The website enquiries had the highest average value, by a wide margin, because they were larger jobs where people had researched first.

No tracking of any kind was involved, and the sheet answered a question their analytics had never been able to.

Tie the two ends together

Once you count outcomes, the aggregate sources become useful in a way they were not before.

If enquiries rise in a month, you can look at which pages gained visits and which searches gained clicks, and form a reasonable view.

That is correlation rather than attribution, and for most small businesses it is enough to make a decision.

The precision people chase, knowing that this specific person came from that specific search, is rarely what changes what they do next.

What this cannot tell you

Being clear about the limits keeps the method honest.

You will not know which of several touches deserves the credit when somebody saw an advert, searched, and later called.

You will not see visitors who considered you and left without contact, except as an absence.

And you will not get the granular funnel reporting that a shop with thousands of transactions genuinely needs.

For a business doing tens or low hundreds of jobs a year, none of those gaps changes a decision. For a high-volume operation, they do.

The counter-case

There is a real argument for proper tracking and it should not be dismissed.

Anybody spending meaningful money on advertising needs conversion data feeding back to the platform, because that is how the campaign learns, and hand counting cannot do it.

Any business with enough volume that a one percent difference matters needs measurement fine enough to detect one percent.

And where several people work on marketing, a shared automated source prevents the arguments that hand-kept records invite.

The point is not that tracking is wrong. It is that a great many businesses install it, never read it, and could have answered their actual question with a spreadsheet and four minutes a week.

The setup

  1. Make one sheet: date, name, source, quoted, won, value.
  2. Fill it in for every enquiry, from any channel.
  3. Ask how they found you and record the answer verbatim.
  4. Check your search console monthly for searches and clicks.
  5. Look at server logs once, with bots filtered.
  6. Review the sheet quarterly, not weekly.

Five months of that beats two years of an analytics account nobody opens.

Collecting your own data is covered in first-party data for a business with none.


Frequently asked questions

What can I measure without tracking visitors?

Everything at the outcome end: form submissions, phone calls, quotes issued, jobs won, and bookings. Plus server logs and search console for aggregate context.

Why start with outcomes rather than visits?

Because outcomes are recorded properly, since money depends on them. Nobody loses an invoice, while plenty of businesses lose a month of analytics without noticing.

Are server logs useful?

Yes, and nothing can block them since they do not run in the browser. The catch is noise: much raw log traffic is automated, so it must be filtered before the numbers mean anything.

Is counting enquiries by hand good enough?

For a business with tens of enquiries a month, yes. It is more accurate than most automated conversion tracking and the person counting notices patterns a report would not show.

What does the search console give me?

Which searches showed your pages, how often they were clicked, and which page appeared. It describes intent rather than behaviour and tracks nobody individually.

When do I actually need tracking?

When you spend meaningfully on advertising and need conversions fed back to the platform, or when volume is high enough that small percentage differences matter.

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.

Analytics account you never open?

One sheet, six columns, four minutes a week. Five months of it will tell you more.

Start a Conversation