Default reporting credits whichever source the customer used immediately before enquiring. That is usually a search for your business name, which means the channel receiving the credit is frequently the one that did the least persuading. The earlier touches that created the demand receive nothing.

What last click does

It assigns a hundred percent of the result to the final interaction and zero to everything before it.

The appeal is that it is unambiguous and requires no assumptions. The problem is that the final interaction is usually the most passive one. By the time someone searches your business name and clicks, the decision has been made. Whatever made it happen occurred earlier and received no credit.

A realistic journey

How a typical service business job actually arrives.

Last click credits the branded search. The guide that started it, the referral that confirmed it, and the vehicle that reinforced it all record nothing. On that reporting, the sensible decision is to cut the content and keep doing whatever produced the branded search, which is precisely backwards.

What gets systematically under-credited

The pattern is consistent: whatever creates demand is under-credited and whatever captures it is over-credited. Since capture is cheaper and easier to measure, budget drifts toward it, and after a year or two the demand it was capturing runs down with no obvious cause.

Why the alternatives do not help much at small scale

Larger organisations use models that distribute credit across touches. For a local business these are mostly unavailable in practice.

They need enough conversions to be statistically meaningful, which a business with fifteen enquiries a month does not have. They only see touches that occurred in a browser, so the vehicle and the neighbour remain invisible. And they cannot connect a research visit on a phone to a call from a landline.

A more sophisticated model applied to incomplete data produces a more confident version of the same wrong answer. That is worse than a simple model everyone knows is limited, which is part of the wider point made in what your analytics cannot tell you.

What actually works for a small business

Ask

"How did you hear about us?" It takes four seconds, it covers the offline half no system can see, and the answers are frequently nothing like the analytics picture. Recorded consistently for three months, it is better attribution than any model available at this scale.

Watch branded search as the demand signal

Rising branded search means something is creating awareness. It does not tell you which thing, and it does tell you that demand creation is working, which last click reporting will never show.

Change one thing at a time

The most reliable method available. Pause a channel for a month and watch total enquiries, not that channel's reported figure. If total enquiries fall, it was contributing more than its attributed share.

Judge the whole rather than the parts

Total enquiries against total spend is a measure nobody can misattribute. It is blunt and it is honest, and it is the number to hold onto when the channel reports disagree with each other.

The decision this protects you from

Cutting the thing that is working because something else took the credit.

It happens in a recognisable way. Content is running, enquiries are healthy, and the reports attribute almost everything to branded and direct traffic. Content looks like it produces nothing, so it stops. Six months later enquiries decline and nobody connects the two, because the decline appears in channels that were never producing the demand in the first place.

Understanding that the last click is a record of where the journey ended, rather than what caused it, is most of what is needed to avoid that.


Frequently asked questions

What is last click attribution?

Assigning the entire result to whatever the customer interacted with immediately before enquiring, and nothing to anything earlier. It is unambiguous and it credits the most passive step in the process.

Why is last click attribution a problem?

The final touch is usually a search for your business name, by which point the decision is made. Whatever created the demand happened earlier and receives no credit, so budget drifts away from it.

Should a small business use a multi-touch model?

Usually not. Those models need enough conversions to be meaningful, only see touches inside a browser, and cannot link a phone visit to a landline call. Applied to incomplete data they produce confident wrong answers.

What is the best attribution method for a service business?

Asking callers how they found you, recorded consistently. It takes four seconds, covers the offline half no system can observe, and after three months it is better than any model available at this scale.

How can I test whether a channel is working?

Pause it for a month and watch total enquiries rather than that channel's own figure. If the total falls, it was contributing more than it was being credited with.

What gets under-credited most often?

Content and guides, reviews, your listing, and anything offline such as vehicles and word of mouth. All do their work early or invisibly, and none is usually the final click.

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.

Not sure which marketing is actually producing work?

We combine what your callers tell you with the search and enquiry data, so demand creation gets credited instead of quietly defunded.

Start a Conversation