Traffic with no enquiries is the most common problem we are called in for, and it is almost never a design problem. In five years of these audits, the cause has fallen into the same five buckets nearly every time. Work through them in this order - the cheap fixes are first for a reason.
1. You are counting leads wrong
Start here, because roughly a third of the time there is no conversion problem at all. There is a tracking problem.
Form submissions that fire no analytics event. Phone calls that never get attributed. A contact form that has been quietly failing since a plugin update. WhatsApp enquiries that land on someone's personal phone and never reach the CRM. We have opened sites where the form had been broken for four months and nobody knew, because the page still said "thank you" afterwards.
What to check today
- Submit your own form. Does the email actually arrive? Check spam.
- Is there a conversion event in analytics, and does its count roughly match your inbox?
- Are phone numbers click-to-call on mobile, and are those clicks tracked?
- Does someone actually own the inbox the form sends to?
A form that silently drops enquiries costs more than any ad budget you are worried about. It is also the fastest thing on this list to fix.
2. The traffic is the wrong traffic
Ten thousand visitors who wanted a definition are worth less than forty who wanted a quote. If your best-performing pages are informational - "what is", "how does", "types of" - you are attracting research intent, not buying intent.
This is not wasted, but it needs a different job. Informational pages build authority and feed the top of the funnel. They should not be judged on direct enquiries. Look at which pages your actual customers landed on before converting, and put your effort there.
3. The page does not answer the three questions
Every visitor arrives with the same three questions, and they leave if the page does not answer them quickly.
- Do you do the specific thing I need? Not "digital solutions" - the actual service, in the words they would use.
- Can I trust you? Real work, real names, real numbers. Stock photography and "20 years of excellence" do the opposite.
- What happens if I contact you? A sales call? A quote? A trap? Say it plainly and fewer people hesitate.
Most pages answer the first, gesture at the second and ignore the third entirely.
People do not abandon forms because the form is long. They abandon because they are not sure what happens next.
4. The site is too slow on the device people actually use
Test on a mid-range Android phone on mobile data, not on your desktop over office wifi. That is the real experience for most visitors, and the gap between the two is enormous.
Speed does not just annoy people. It removes them before they see anything. A page that takes four seconds to become usable has already lost a meaningful share of the visitors you paid to get there - and search engines factor the same signals into ranking, so you lose twice.
The usual culprits, in order of how often we find them: uncompressed hero images, four separate tracking scripts loading in the head, a page builder shipping CSS for components the page does not use, and fonts that block rendering.
5. You are asking for too much, too early
A nine-field form that demands company size and budget from someone who has known you for eleven seconds converts badly. Not because people mind typing, but because the ask is disproportionate to the relationship.
Two options that both work: shorten the form to name, email and one open question, and qualify afterwards by phone. Or keep the long form but offer a lower-commitment alternative alongside it - a callback request, a WhatsApp link, a direct number.
A quick benchmark
For most service businesses, one to three per cent of visitors making an enquiry is normal. Under one per cent, something on this list is wrong. Above five per cent, either your traffic is unusually well-qualified or you are counting something that is not a lead.
How to run this diagnosis in an afternoon
You do not need tooling beyond what you already have. Two to three hours, in this order.
Hour one: prove the plumbing works
- Submit your own contact form with a real email. Time how long the reply takes to arrive, and check spam.
- Call your own tracked number from a phone that is not yours. Confirm it rings and the call is recorded in whatever system you use.
- Open analytics and compare recorded conversions against your inbox for the last thirty days. If they disagree by more than a little, stop and fix that first.
- Check your form's destination mailbox is one a person actually reads.
Hour two: read the traffic honestly
- List your ten highest-traffic landing pages.
- Beside each, write what the visitor was probably trying to do.
- Mark which of those intentions could plausibly end in an enquiry this month.
- If most are marked no, your problem is traffic mix, not the page.
Hour three: use your own site as a stranger
- On a phone, on mobile data, open your top landing page.
- Count the seconds until you can read something useful.
- Ask yourself the three questions - do they do my thing, can I trust them, what happens if I make contact - and note where each is answered, if it is.
- Try to make an enquiry without using the main navigation.
Most people find at least two of the five problems this way, without spending anything.
The things people fix that rarely help
Ranked by how often we see them attempted before anything on the real list.
- A full redesign. Expensive, slow, and it changes everything at once so you learn nothing about what mattered.
- A live chat widget. Sometimes helps. Often adds another inbox nobody watches, and a second later-loading script.
- More blog posts. Useful over a year. Does nothing for enquiries this quarter.
- Colour and button changes. The classic A/B test. Real effects exist but they are small, and they are invisible under the noise of a bigger problem.
- More ad spend. Pouring more traffic into a page that does not convert simply raises the cost of learning that.
Nothing you spend on traffic matters until the thing traffic lands on works.
Two patterns worth knowing
The seasonal panic
Enquiries drop, everyone assumes the site broke, and a rebuild gets commissioned. Check the same month last year before doing anything. Trades, education, property and travel all have swings large enough to look like failure. A year-on-year comparison takes two minutes and has saved several of our clients a rebuild.
The invisible form regression
A plugin updates, a mail provider changes a rule, someone edits an email address with a typo. The form still shows a success message because that message is client-side and has no idea whether anything sent. Enquiries drop to zero and nothing looks broken.
The fix is boring and worth doing: a form that writes every submission to a file or database before it tries to email, plus a monthly test submission in someone's calendar. We build ours that way, because email delivery is the single least reliable part of any contact form.
What good looks like afterwards
You should end up able to answer four questions without asking anybody: how many enquiries came in last month, what each of them said, which pages produced them, and what each one cost. If you cannot answer all four, the measurement work is not finished, and everything built on top of it is guesswork.
The order matters
Work top to bottom. Tracking first, because everything after it is guesswork without numbers you trust. Then traffic quality, because no amount of copywriting fixes the wrong audience. Then the page itself. Then speed. Then the form.
Most agencies start at step three, because redesigns are easier to sell than tracking audits. That is why so many redesigns change nothing.
If you would rather not work through it alone, we run this as a fixed-scope conversion audit and hand you the findings in writing. What we fix, and what we deliberately leave alone, is part of the same document. You can also read how we set up reporting so this never becomes invisible again.