Ask most business owners which of their marketing brings in customers and you get an impression rather than an answer. Instagram, probably. Word of mouth. Somebody mentioned they saw us somewhere.
The impression is usually wrong in one specific direction: it over credits whatever is most visible to you and under credits whatever is quiet. Search is nearly always the quiet one, because a customer who found you on Google and then messaged you on WhatsApp looks identical to a customer who was referred by a friend.
Three things fix this. None of them requires expensive software.
One. Ask, and write it down
The simplest answer is the one businesses skip because it feels unscientific.
Add one question to whatever you already do with a new enquiry: how did you hear about us. Then record the answer somewhere consistent, a notebook, a spreadsheet, a column in whatever you already use. Not in your head.
People misremember, so this is imperfect. It is also the only method that captures the customer who saw your flyer, asked a friend, then searched your name a week later, which is a very common path and one no software will ever attribute correctly.
After two months of consistent recording you will know more about your own demand than most businesses in your market.
Two. Make the website capable of reporting
If you have a website and no analytics on it, you are running the one channel that can be measured precisely and choosing not to measure it.
The setup is free. Google Analytics tells you how many people arrived, what they looked at, how long they stayed and roughly where they came from. Google Search Console tells you something Analytics cannot: which searches showed your site, how often people clicked, and which pages they landed on. The second one is the more useful of the two for most businesses, and it is the one most have never opened.
Two things to be careful about, because they quietly ruin the data.
Exclude yourself. If you and your team load your own website all day, your traffic is in the numbers, and for a small site that is a large distortion.
Check it is actually installed and reporting. A tag added incorrectly fails silently. The site looks completely normal and no data ever arrives, which is the worst possible failure, because you think you are measuring and you are not.
Three. Tag your links
This is the part that separates guessing from knowing, and it takes minutes.
When you put your website address in your Instagram bio, in a WhatsApp broadcast, on a flyer, or in an email, add a label to the end of the link. It looks like this:
yoursite.com/?utm_source=instagram
Anyone clicking that link is recorded as having come from Instagram, no matter how they got there. Use a different label for each place: instagram, whatsapp, flyer, email, linkedin. It is the same website, it just arrives labelled.
For a printed flyer, where nobody types a long link, point a QR code at the tagged address. The customer sees nothing different and you find out the flyer worked.
Do this consistently for a month and the question of which channel is working stops being a matter of opinion.
What you will probably discover
Three things come up repeatedly when businesses start measuring.
Search is bigger than they thought, because search traffic is invisible until you look. Social is smaller than they thought, because the activity is visible even when the outcome is not. And a large share of enquiries arrive through a route nobody had considered, often a single page, a single listing, or one person who keeps referring people.
That last one is the most valuable finding, because it is usually something you could do more of deliberately and have been doing accidentally.
Start here if you do nothing else
Ask every new enquiry how they found you and write it down. Open Search Console and look at what your site is already being found for. Tag the link in your Instagram bio.
That is an hour of work and it changes the next twelve months of spending from a guess into a decision.
Our free deep digital audit checks whether your website is measuring anything at all, and whether what it is measuring can be trusted, entirely from the outside.