Your website is not converting: where to start

State the problem before solving it

"The site does not convert" covers three very different situations that call for different answers: the site gets few visitors; it gets visitors but the wrong ones; or it gets the right ones and they leave without doing anything.

The three separate in minutes. Look at how many visitors arrive, where from, and on which pages. A site with thirty visitors a month does not have a conversion problem: it has a visibility problem, and rebuilding it will not help.

What foreign guides miss: in Morocco, people do not fill in the form

Most conversion articles start from an implicit assumption: an interested visitor fills in a form. In the Moroccan market that assumption rarely holds. A hesitant buyer calls, or messages on WhatsApp. The form is the channel they use least, and often the one given the most prominence.

The practical consequence is immediate. If your only conversion measure is forms received, you are measuring the least-used channel and concluding the site does not work — when it may be producing calls nobody counts.

Before any other change: show a tappable phone number and a WhatsApp link, and count them. A number as an image, or as plain unlinked text, is friction for free on a device where nobody retypes a number.

The order to look in

Each step below is cheaper than the one after it. Work down in order, and stop as soon as you find something.

• Mobile speed, on a real network. Not on your desk over fibre. A page that takes several seconds to show anything loses visitors before it is read.
• What the first screen says. A visitor should understand what you do, for whom, and where, without scrolling. Many sites put a slogan and an image there; the visitor is looking for an answer.
• The number of form fields. Every additional field costs responses. Name, contact, need: three fields are almost always enough for a first exchange. Job title, budget and address can wait for the second message.
• The button label. "Send" does not say what happens next. "Request a quote" or "Get a callback" describes the outcome, and asks for less.
• Trust signals. An address, a number that answers, visible work, faces. In a market where a lot is still bought on trust, an anonymous business is a risky one.
• What happens after submission. An enquiry answered three days later is a lost enquiry. The site is no longer the problem at that point, but it cancels everything else.

The expensive wrong turns

Three reflexes recur, and cost a lot for little effect.

Redesigning first. A full rebuild sometimes fixes the problem by accident, correcting speed and structure along the way. That is an expensive way to find out which of the two mattered.

Adding traffic to a site that does not convert. Doubling visits also doubles the number of people who leave. Fix first, buy traffic after.

Adding features. A chat widget, a configurator, a booking tool. Useful once the journey works; noise while the page still does not say what you do.

Measuring without building a machine

The diagnosis above assumes numbers. The good news is that very few are needed, and they can be set up in an hour.

Make each channel traceable. A distinct WhatsApp link per main page, a tappable phone number, and the form. Three counters are enough to know which one is working.

Count enquiries, not visits. A spreadsheet with one row per enquiry, the date, the channel and the outcome tells you more than any dashboard. After two months you know which channel brings customers and which brings noise.

Watch a recorded session if you can. Ten minutes observing real visitors replaces weeks of guessing — and where they stop is never where anyone expected.

Frequently asked

And at Noor Studio Agency?

We measure before proposing anything: where visitors come from, what they do, and which channel they actually use to reach you. A rebuild decided without those three answers is a bet.