Short answer

A lack of enquiries can result from poorly matched traffic, an unclear offer, difficult contact options or faulty measurement. First confirm that messages really are not arriving. Then follow the complete journey from a specific traffic source to receipt of the enquiry. Only a verified problem should define the scope of a paid website upgrade.

First distinguish a lack of contacts from a lack of data

Page views do not tell you how many people are considering a purchase. The same user may open several pages, and some visits may concern a guide, recruitment or information for existing customers. Compare similar periods, traffic sources and landing pages. Account for seasonality and the length of the buying decision; do not draw firm conclusions from a handful of visits.

This is a diagnostic guide, not a promise of sales growth. Without access to analytics, the inbox and the enquiry-handling process, it is not possible to identify one cause fairly. The order below is a recommendation: eliminate failures and measurement errors first, then assess content and design.

1. Does the traffic come from potential customers?

Separate visits from advertising, search, referrals and direct access. Check whether visitors want a service you actually provide, in the right region. A popular educational article may attract people who want to solve a problem themselves rather than hire someone. That is not evidence that the whole website is ineffective.

For the most important landing page, record the source, the promise preceding the click and the expected next step. If an advert concerns one service but leads to a general business presentation, start by aligning the message and destination.

2. Does the first screen explain the offer?

Open the website without prior context. Is it clear what the business does, for whom and within what scope? “Comprehensive business solutions” does not answer these questions. Ask someone outside the company to describe the service in their own words after reading the first screen. Record their answer instead of prompting them.

If the offer needs an oral explanation, improve the specific heading and short description. A new animation or larger photo cannot replace information needed for a decision. Do not add promises or deadlines the business cannot meet.

3. Can the customer assess the scope of working together?

Check whether the description explains what the service includes, excludes and how the engagement begins. No public price is not necessarily a problem, but no explanation of how pricing works increases uncertainty. Explain what information is needed for a proposal and what the customer will receive in response.

Do not invent a price list merely to fill a gap. For a custom-priced service, an explanation of cost factors, stages and limits is often more useful than an arbitrary “from” amount.

4. Is the evidence specific and credible?

General quality claims do not replace examples of work. Check whether projects describe the problem, delivered scope and how the result was verified. Publish results only where data and permission to use it exist. If you only have project photos, present them honestly as a portfolio, not evidence of sales growth.

Also check basic business information and whether contact details are current. Customers should not have to establish for themselves who is responsible for the service. A lack of case studies does not justify invented recommendations.

5. Is the next step clear?

On a specific service page, choose one primary goal: an enquiry, booking or conversation. Several equally prominent buttons can make choosing harder, especially if each leads elsewhere. A button label should describe the action, not merely say “more”.

Follow the link with a normal click and inspect the final screen. A changed URL or fragment after # does not prove the form is visible. After clicking, users should not have to search the whole website for contact options again.

6. Can a customer comfortably complete the contact journey on a phone?

Check the website on a real phone: navigation, contact button, form fields and on-screen keyboard. Look for overlapping bars, consent banners, invisible messages and elements extending beyond the screen. No horizontal scrolling is not enough if a button covers another field.

Test recovery after an error too. Does entered information remain? Does the phone number trigger the correct action? Focus on the whole journey, not only an attractive opening screen.

7. Does the form ask only for necessary information?

For every required field, ask whether a conversation really cannot begin without that information. Do not remove requirements needed to handle the enquiry, but explain those that may cause uncertainty. Label fields and provide instructions before an error occurs.

W3C recommends labels that identify form controls and are programmatically associated with them. Text that disappears when typing begins should not replace a clear instruction. This concerns accessibility and understanding, not a guaranteed conversion improvement.

8. Does the message actually reach the person handling enquiries?

A “sent” message does not confirm receipt by the right person. In an agreed test, check the journey to the inbox or CRM, the spam folder and the person responsible for replying. Label the test so it is not reported as a genuine lead. Do not send customer test data without authorisation.

If enquiries arrive but go unanswered, the problem extends beyond the website. Record who owns the process and how responses are handled. A website rebuild will not fix a process nobody manages.

9. Does performance make the task difficult?

Assess loading, button responses and layout movement. The web.dev documentation distinguishes these aspects in Core Web Vitals: LCP, INP and CLS. A laboratory score is a diagnostic clue; compare it with real-user data where available.

Do not commission optimisation solely to change a score’s colour. Name a specific fault in the brief, such as a moving button or delayed form opening, and agree an acceptance test. A good technical score does not prove that the offer fits the audience.

10. Does measurement distinguish a click from an enquiry?

Google Analytics describes generate_lead as an event related to submitting a form or request for information. Clicking a contact button is an earlier step and should not masquerade as a completed enquiry. Check whether the event fires despite a sending error or repeatedly after refreshing.

Compare measurement with enquiries actually received, accounting for consent limits, tracking blockers and phone contacts. Analytics is not a complete sales register. Record enquiry quality too: more irrelevant contacts do not necessarily mean a better result.

A targeted fix or a rebuild? Set the scope using evidence

Confirmed problemInitial work scopeAcceptance criterion
A form or delivery failureRepair the contact journeyA labelled test reaches the correct recipient
An unclear offer on sound technologyImprove one service’s content and hierarchyThe visitor understands the scope and next step
Problems repeat across multiple templatesAudit and upgrade planA fault list, priorities and tests for each template
No reliable measurementAgree definitions and check eventsA click and a received enquiry are distinguished

Before commissioning work, prepare the affected page URL, traffic source, observation and evidence. Separate the fact “a button covers a field” from the hypothesis “customers do not trust the business”. Expect scope, tests and a before-and-after comparison method from the provider, not a promise of a specific enquiry count without data.

Sources and limits of the conclusions

Sources verified on 4 November 2026. The diagnostic order and scope matrix are editorial recommendations. They do not present a particular client’s results.

Need to define the scope of website changes?

Prepare your current website URL and describe the observed problem in the website brief. This starts a discussion about scope; it is not an automatic diagnosis or a guarantee of enquiries.

Open the website brief (in Polish) ↗