Before I change anything on a site that brings in no enquiries, I check one thing - whether the problem sits in the traffic or in the site itself. That sounds obvious, but it decides everything that follows, because low conversion on a company website comes from two completely different families of causes, and the cure for one does not move the other at all. A site with no traffic has nothing to convert. A site with traffic that collects no enquiries has a specific obstacle inside it, and that obstacle can be found.
The most common mistake I see goes like this: the company feels that “the website is not working”, so it orders a rebuild. Three months later a new site is live, better looking, and the phone rings exactly as often as before. The cause was somewhere else, and nobody checked it before the money was spent.
Below I go through the causes in the order I actually check them on a client’s site - starting with the ones that break measurement and disguise themselves as everything else, ending with the ones you can see at a glance. For each one I say how to confirm it is the culprit before you change anything.
Start with measurement, or the whole list is guesswork
The first question is not “what should I fix” but “how do you know it is bad”. Surprisingly often, nobody does. The form sends an email, the email lands in an inbox, somebody replies to it, and nothing anywhere is counted. The company judges its conversion from memory, and memory keeps the lean months much better than the good ones.
Before you decide the site converts badly, you need three numbers for the same period: how many people arrived, how many reached the contact page or the form, how many sent it. Only those three side by side tell you where people drop out. If you only have the first one, you do not have a conversion problem yet, you have a measurement problem, and that is the first thing to fix.
The second version of the same problem is worse, because it looks innocent: the measurement exists but lies. It happens that the form submission event was never wired up, that the consent banner blocks analytics for everyone who did not click it, or that the redirect after submitting gets lost on the way. The report then shows zero enquiries next to an inbox full of them.
What helps: send your own form from your phone, on mobile data, and check three things at once - whether the email arrived, whether the enquiry shows up in analytics, and whether the thank-you screen appeared at all. Five minutes, and it decides whether the rest of the list is worth reading.
The traffic does not match what you sell
If the measurement is healthy and there are still no enquiries, the next question is who is actually arriving. A company website can have decent traffic made up entirely of people looking for something else - a definition, a free template, a job, or instructions on how to do it themselves. That traffic will never turn into an enquiry, because none of those people came to hire anybody.
The simplest check is the queries people arrive on and which pages collect the visits. If all the traffic sits on the blog while the service pages stay empty, the problem is not conversion but the shape of your visibility - I wrote about setting that up from scratch in how to start SEO for a business website.
There is a reverse version too: the traffic is right but lands in the wrong place. Someone searches for a specific service, arrives on the homepage, and the homepage talks about the company in general and leaves them to find the rest themselves. Half of them will not look.
The offer does not say who it is for
This is the most frequent cause I find on sites that are technically fine. The headline at the top talks about the company (“comprehensive solutions”, “over twenty years of experience”) rather than about the reader and their problem. Someone who arrived with a specific need has a few seconds to decide whether they are in the right place. If the page does not make that easy, they go back to the search results.
The other side of the same coin is an offer “for everyone”. The company is afraid to narrow down, so it writes that it does everything for everybody. The effect is the opposite of the intention - nobody feels the offer is meant for them. Narrowing scares some people off, but the ones who stay are much closer to sending an enquiry.
The test is simple and unpleasant: show your homepage to somebody outside the company for ten seconds, take the screen away, and ask what the company does and who for. If they cannot answer, neither could your customer. I collected more on the difference between a brochure and a site that brings in enquiries in what a B2B website needs to generate leads.
The form turns an enquiry into paperwork
The form is where you lose people who had already decided, which is why it hurts most. A few things break it regularly:
- Too many fields. Every field is another chance to give up. A first contact usually needs a way to reply and room to describe the problem - the rest is a conversation, not a form.
- Required fields that need not be. A tax number, company name and address at the “I want to ask about a quote” stage is asking for data people are not ready to hand over.
- No word on what happens next. One sentence about when a reply will come does more than redesigning the whole section.
- A form that quietly fails. A plugin after an update, outgoing mail configuration, a spam filter - an outage looks exactly like a lack of interest, except it is free to fix.
- A single route of contact. Some people will never fill in a form but will call or email if the number and address are visible rather than buried in the footer.
What helps: once a quarter, send your own form like a stranger and count how many fields you have to fill in to ask one question. If it makes you impatient, your customer is more impatient still.
The site makes people wait, most of all on a phone
Speed is not a separate topic from conversion - it is the same matter seen from another side. Someone waiting for a page to load drops out before they see the offer, so every point above stops mattering. The clear majority of traffic on the sites I look at comes from phones, and it is on a phone and a weaker connection that a slow site hurts most.
It is worth separating two things here. The first is the time to the first visible content - how long before anything shows up. The second is layout stability: a page where buttons jump around while it loads can be more infuriating than one that is simply slow. Both are measured with the same tools I described in what Core Web Vitals are and how to improve them.
A practical note: do not measure speed on your own computer, on your own network, with a warm cache. Measure on a phone, on mobile data, on a service page rather than only on the homepage.
No proof that anyone took the risk first
The last family of causes is the least technical and the most often skipped. A site can be fast, clear and carry a short form, and still fail to convince, because it gives no reason to trust. Asking for a quote is a small risk for the customer - they spend time and hand their contact details to somebody they do not know.
The things that reduce that risk are boring and effective: specific projects named and described by the problem they solved, rather than an unlabelled gallery; the face and name of the person they will be talking to; clear terms of working together; visible company details. This is not about collecting badges, it is about a human being and finished work being visible on the page.
There is also a point of honesty here: if the site promises more than the company delivers, high conversion ends badly anyway - in conversations that lead nowhere.
What order to fix it in
Order matters, because some fixes mask the effect of others. I keep to the same one every time: measurement first, so the results are visible at all. Then whatever blocks people who have already decided - the form and the routes of contact, because the effect shows up fastest and costs least. Then the clarity of the offer on the service pages. Speed and a wider rebuild come last, because that is the expensive part, and without the three earlier steps it is hard to judge whether it paid off.
If you want to go through this on your own site in that order, reviewing and repairing an existing site is exactly how I work - with a list of what I found before anything gets changed. A wider list of things to check yourself is in the post on a WordPress site audit.
How long should you wait before judging whether a fix worked? Long enough to collect a comparable number of visits, not as long as impatience suggests. With the modest traffic of a company website, two days settle nothing - a few enquiries either way is ordinary chance. Compare full months, and always against the same period, because seasonality can swallow the whole effect of a change or pretend there was one.
Sitting on this yourself and unsure where to start? I speed up and repair existing sites and stores - Core Web Vitals, outages, the features that are missing. Tell me what is happening and you will get a concrete quote with a date.