The short answer
A conversion funnel is the planned path from the first visit to the enquiry or purchase. It has four stages: attention, interest, decision and action. Each stage answers the question the visitor has right now and leads to the next step.
The biggest lever usually isn't more traffic, but the break points: too many equal buttons, forms that are too long, no intermediate step for the undecided. The worked example below shows why a better form often brings more than twice as many visitors.
What is a conversion funnel?
A funnel describes the path a visitor travels from first contact with your website to getting in touch. Many arrive at the top, few enquire at the bottom. That's normal. What matters is how many you lose unnecessarily at each stage.
Most websites have traffic but no deliberate funnel. Visitors land somewhere, read something and leave again without any next step having been planned. A working funnel isn't a coincidence, it's structure.
Which stages does a visitor go through?
1. Attention: am I in the right place?
The visitor arrives via Google, a social media post or a recommendation. They don't know exactly what to expect yet. Now instant clarity counts: what do you offer, for whom, in which region? Anyone who only finds out after three paragraphs is often already gone.
2. Interest: can they really do it?
The visitor stays and wants to know more. They click on service pages, case studies and the about page. Here interest turns into trust, through concrete results, real references and comprehensible expertise.
3. Decision: what does it cost, how does it work?
The visitor is considering getting in touch but has open questions about price, process and duration. Pages that answer these questions in advance noticeably lower the hurdle. An FAQ block or a transparent price frame often works more strongly here than any button. Every unanswered question is a reason to leave the page.
4. Action: how easy is the last step?
Form, call, booking: this is where a visitor becomes a lead. This stage is decided by details. How many required fields are there? Does a confirmation arrive immediately? Is there a low-threshold alternative, such as a short call instead of a long form?
Where does a funnel lose most visitors?
These four break points are the ones we see most often in audits:
- Too many equal calls to action. If newsletter, download, demo and contact are equally prominent side by side, the visitor often chooses nothing at all.
- No intermediate step for early phases. Not everyone is ready for "Call now". A checklist, a self-check or a free audit keeps prospects in the funnel who aren't ready to buy yet.
- Long, generic forms. Only ask what's necessary for first contact. Name, email and a free-text field are often enough. The rest comes in the conversation. In our project experience, every required field beyond three costs completions.
- CTA before trust. A big "Buy now" button before the visitor knows who you are comes too early and gets ignored. The button can be visible early. But it needs evidence directly below it.
An example from our work: at Hotel Zuckerfabrik the availability check used to work only via a detour to an external platform. That's a classic break point at the action stage. After the relaunch the availability widget sits directly in the hero, next to "Book now".
Which next step suits which stage?
Not every visitor is equally far along. Someone still comparing won't click "Book an appointment". That's why each stage needs an offer that matches its question:
| Stage & question | Matching next step |
|---|---|
| Attention Am I in the right place? | Clear headline, link to the matching service page |
| Interest Can they really do it? | Case study, guide article, customer testimonial |
| Decision What does it cost, how does it work? | Price frame, process in steps, FAQ, free audit |
| Action How easy is getting in touch? | Short form, phone, 30-minute call |
The main button stays the same at every stage. The intermediate steps are quieter, for example a text link or a box at the end of a section. That way you don't lose the undecided without distracting the decided.
Worked example: more traffic or a better funnel?
Example (fictional but typical figures): a service provider has 2,000 visits a month. The table shows how many people remain at each stage.
| Stage | Today | Better form |
|---|---|---|
| Visits | 2,000 | 2,000 |
| Service page read (30 %) | 600 | 600 |
| Form opened (10 %) | 60 | 60 |
| Form submitted (40 % → 60 %) | 24 | 36 |
Today 24 enquiries a month come in, that's 1.2 % of visits. If the form becomes shorter and clearer, so that 60 instead of 40 % submit it, it's 36 enquiries. That's 50 % more, without a single extra visitor.
To get the same result via traffic you'd need 3,000 instead of 2,000 visits, i.e. 50 % more reach. That costs ad budget or months of SEO work. That's why looking at the funnel usually pays off before investing in more visibility. Conversely: if hardly any visitors come, the best funnel won't help. Then SEO & GEO is the first step.
What does a funnel look like on a service page?
A proven basic pattern for a single service page:
- Hero: service, target group, one button. Answers "Am I in the right place?"
- Services and results: what you do concretely and what comes out of it.
- Evidence: cases, figures with a time period, a real customer testimonial.
- Objections: FAQ, price frame and an honest "when we're not a fit".
- Close: the same call to action as at the top, plus phone and email for everyone who doesn't like forms.
Every section has a job in the funnel. None is just decoration. If you can't name a visitor question a section answers, it can probably go.
How do you measure your funnel?
Without measurement, every improvement remains a gut feeling. With Google Analytics 4 it takes two steps. First you mark the important actions as key events, such as submitting the form [2]. Then you set up a funnel exploration. It shows step by step how many users get further and where they drop off [1].
Two tips from practice. With a consent banner you only see the visitors who agreed. So the figures are a sample, not a complete count. And always compare equal periods, for example four weeks before and four weeks after a change. Change only one thing at a time if possible, otherwise you won't know what worked.
What happens after the enquiry?
The funnel doesn't end with sending. An immediate confirmation by email shows that the enquiry has arrived and what happens next. Anyone who signed up for an intermediate step such as a checklist needs a short sequence of emails that keeps accompanying them. Which flows pay off is in the article on automation & sequences.
In B2B there's another step: the enquiry lands in the CRM, and sales needs to know who comes first. How data, signals and automation work together here is explained in the article What is GTM engineering?.
Self-test in five questions
- Can you see in analytics at which point most visitors drop off?
- How many different CTAs does your most important page have? More than two are usually too many.
- How many required fields does your contact form have? Every one beyond three needs a good reason.
- Is there an intermediate step for visitors who don't want to enquire yet?
- Can you say for every section which question it answers and where it leads?
A deliberately designed funnel is often the biggest lever for turning existing traffic into more enquiries. Often it doesn't even take more visitors.