Website traffic but no leads? A 15-minute diagnosis

Which of the four problems do you have?

Answer these four questions in order. Stop at the first "no": that's where your leads are going.

1. Are the visitors the kind of people who buy from you? If not, you have a traffic problem, not a website problem.
2. Can a ready visitor see what to do next, on the page they're reading? If not, they leave because nothing asks them to stay.
3. Does your contact route work, and is it quick to use? If not, people try and give up, or their message never reaches you.
4. Are your visitors ready to talk yet? If they're early in their search, they need something lighter than "contact us".

The order matters. A better form won't help if the visitors came looking for something you don't sell. A smarter offer for early researchers won't help if the form behind it is broken.

	What you see	The check	The fix
1. Wrong traffic	Lots of visits to one blog post, short visits, visitors from countries you don't serve	Search Console: the searches and pages that bring people in	Write for buyer searches; send ads to matching pages
2. No next step	Visitors read your service and pricing pages, then leave	Open your top 5 pages: what does a ready visitor click?	One clear, quick next step on every money page
3. Broken or slow form	Few submissions, or none for weeks	Send yourself a test from your phone; count the fields	Fix delivery; ask only what you need now
4. Not ready yet	Return visits, long reads, no contact	Look for repeat visitors who read but never ask	A lighter step for later: a guide, a price list, a reminder

Minutes 0–4: are they the right visitors?

Start here, because everything else depends on it. Open Google Search Console, go to the Performance report, and look at the Queries tab and then the Pages tab. They show the searches people typed to find you and the pages they landed on.

Ask three things:

• Do the searches sound like buyers? "Emergency plumber Ghent" is a buyer. "How does a boiler work" is a student, or someone who'll fix it themselves. Both are fine visitors, but only one of them was ever going to call you.
• Where does most of the traffic land? A common pattern: one old blog post brings in half of all visits, from people who will never need your service. Your total looks healthy, and your service pages get a trickle.
• Do the places match? A local business with most visits from other countries has a traffic number, not a market.

Then check your other sources in your analytics: ads, social posts, newsletters. They don't convert equally. In Unbounce's 2024 benchmark of more than 41,000 landing pages, visitors from email converted 60% more than visitors from paid social, 77% more than paid search and 370% more than display ads. A visitor who already knows you is worth several who clicked a banner.

What to do if this is your problem: take your traffic numbers with a pinch of salt and look only at visits to pages that sell something. Write the next few articles for searches a buyer would type ("cost of", "near me", "vs", "for small business"). If you run ads, send each ad to the page that matches what it promised, not to your home page.

If the traffic is right, move on.

Minutes 4–8: is there a next step for someone who's ready?

Open the five pages that get the most visits from the right people. Usually that's the home page, a service page or two, pricing, and maybe a case study. On each one, pretend you've decided to buy and look for what to do.

Many small business sites fail this check in one of three ways:

• The only step is at the very bottom, or in the menu as "Contact". Someone who's ready halfway down the pricing page has to go looking.
• The step is bigger than the decision. "Request a proposal" asks a lot from someone who only wants to know if you're free next month.
• There's no answer to the question that stops people. Price, availability, area served, how long it takes. If the page doesn't say, the visitor has to ask, and many won't.

A rule of thumb from the landing page data: clear, simple pages do better. In the same Unbounce study, pages written at a 5th to 7th grade reading level converted at 11.1%, more than twice the 5.3% of pages written at a professional level. Short words and short sentences aren't dumbing down. They're how busy people read.

What to do if this is your problem: put one next step on every page that sells, close to the point where people decide. Make it match the moment: "Call now" for an emergency service, "Book a 15-minute call" for an agency, "See if we're free in your area" for a trade. Answer the blocking question (price range, area, timing) on the page itself. For more on spotting the visitors who are ready and making it easy for them, see how to get hot leads from your website.

Minutes 8–12: does the form work, and is it worth filling in?

Now test the route itself. Take your phone, open your contact page, and send yourself a message the way a customer would. Then check three things.

• Did it arrive? Forms break quietly. A plugin update, a changed email password, a spam filter. If the last enquiry was weeks ago, this is the first suspect. Check your spam folder too.
• How did it feel on the phone? Tiny fields, a dropdown with 40 options, a captcha you had to solve twice, an error that wiped what you typed.
• What happened after you sent it? A clear "thanks, we'll call you today" or a blank page?

Then look at what the form asks. Nielsen Norman Group's first rule for web forms is to keep them short: cut any field you can work out yourself, ask later, or do without. Their article also cites a study where forms that followed their guidelines were submitted without errors on the first try 78% of the time, against 42% for forms that didn't.

But don't expect a shorter form to fix everything. Zuko, which measures forms on many websites, found that 34% of people who start a form don't finish it, and that the length of a form makes almost no difference to how often people complete it. The bigger loss comes earlier: in Zuko's data, only 9.09% of people who see a contact form submit it, the lowest rate of any type of form they track. Most people never start.

That points to the real job of a form: it has to feel worth it. A blank "Message" box asks the visitor to do the work. A form that asks one useful thing ("What do you need help with?" with answers to tap) feels like the start of a conversation.

What to do if this is your problem: fix delivery first and send a test every month. Then cut the form to what you need for the first reply, usually a name, one way to reach them and what they need. Offer a quicker route next to it for people who'd rather not type: a phone number, WhatsApp, or a call back at a time they choose. For what to ask, see lead qualification without long forms, and for booking forms that ask questions first, Calendly routing forms vs asking one question.

Minutes 12–15: are they just not ready yet?

If the traffic is right, the next step is clear and the form works, the most likely answer is the simplest one: most of your visitors are still looking.

The research on business buyers is clear on this:

• They contact you late. In 6sense's 2025 study of nearly 4,000 B2B buyers, buyers first contacted a seller when they were 61% of the way through their buying process.
• They come to you, not the other way round. In the same study, 79% of buyers said they started the contact themselves, and 94% had ranked their shortlist before talking to any vendor. They bought from their top-ranked vendor 77% of the time.
• Many would rather not talk to sales at all. In a Gartner survey of 632 B2B buyers, 61% said they prefer a buying experience without a sales rep. They do want a seller's view on one thing: whether the product or service fits their needs.

So a visitor who reads three pages and leaves isn't necessarily lost. They may be building the shortlist you want to be on top of. What they need from you right now isn't a sales call. It's the information that makes you their first choice, and an easy way back when they're ready.

You can see these visitors in your analytics: returning visitors, long visits to service and pricing pages, several visits over a week or two with no contact.

What to do if this is your problem: offer a smaller step for people who are still looking. A price list or a "what it costs" page. A short guide that answers what they'll ask anyway. "Email me the quote checklist." A reminder: "Want us to email you when we have a slot in March?" Ask for an email only, and only with their consent. And keep the bigger step there for the day they come back ready.

A worked example

This is an illustration, not a customer story. A three-person accounting firm gets about 1,500 visits a month and two or three enquiries. Here's how the 15 minutes go.

1. Traffic (minute 3): Search Console shows that about half the visits land on an old article about filling in a tax form, from people doing their own taxes. That traffic isn't wrong, but it was never going to hire an accountant. The pages that matter, the services and prices, get around 400 visits.
2. Next step (minute 6): The services page ends with "Contact us" and a link to a form. The price of the most popular package isn't on the page.
3. Form (minute 10): The test message arrives. The form asks for nine things, including company number and annual turnover.
4. Readiness (minute 14): Many of the 400 come back two or three times in the weeks before a tax deadline, read the prices page and leave.

The diagnosis: the 1,500 was never the real number; 400 is. Of those, many are comparing firms ahead of a deadline. The firm adds a price range to the services page, cuts the form to three fields, adds a "call me back" option, and puts a short question on the pricing page for returning visitors: "When do you need this done?"

• "This month": offer a call back today.
• "Before the next deadline": offer the checklist of documents to prepare, by email, and a reminder two weeks before.
• "Just comparing": show what's included in each package. No more questions.

Where Signal fits

Problems 2 and 4 are what we built Signal for. A small script on your site reads each visit: the pages read, where the visitor came from, how long they stay, and whether they've been before. When the visit hasn't made it clear what someone needs and the answer would change what to offer, it asks one of your questions, with answers to tap. Then it offers the next step you wrote for that answer: a call, a call back, WhatsApp, or something lighter for people who are still looking. A visitor who's ready can leave a phone number or an email in one field, and your team hears about it with the whole visit.

It won't fix problems 1 and 3. If the traffic is wrong or your form is broken, start there; no tool on the page will help. Signal never stores IP addresses, and you can run it in shadow mode first: it shows what it would have done on your real traffic, without visitors seeing anything.

When this diagnosis isn't enough

• You sell online and want orders, not enquiries. The checks are similar, but the details differ: basket, delivery costs, returns. See why no one is buying from your Shopify store, and for the season, Black Friday traffic with no sales.
• Your traffic is very small. With a few visitors a day, the numbers in these checks won't tell you much. Ask your last ten customers how they found you and what nearly stopped them.
• Leads come in, but they're the wrong ones. That's a qualification problem, not a traffic one. Ask what they need and how soon before you book a call; lead qualification covers how.
• You want names, not leads. Tools that look up who visits your site exist, but they have limits and legal questions in the EU. See whether website visitor identification is GDPR compliant before you buy one.

Questions people ask

What's a normal conversion rate for a website? It varies a lot by industry and by page. In Unbounce's 2024 data, the median landing page converted 6.6% of visitors, with industry medians from 3.8% to 12.3%. A whole website, with blog posts and job pages, will be lower. Compare your service pages with each other over time, not with a benchmark.

Should I just buy more traffic? Not before the four checks. More of the wrong visitors, or more of the right ones arriving at a page with no next step, costs money and changes nothing.

How often should I run this? Test the form every month. Run the full 15 minutes every quarter, and whenever leads drop without a reason you can see.