Traffic is useful only when it helps suitable people move towards a business outcome. If visits are rising but enquiries remain flat, buying more traffic may make the problem more expensive. Diagnose the visitor journey, success event and lead quality before changing the site.
Confirm that “no leads” is a real measurement finding
Complete the form, call, booking or checkout yourself and verify the success state. Check whether analytics records the event once, whether spam filtering or CRM routing loses submissions and whether phone or messaging enquiries are counted separately. Compare leads with sessions using the same date range and filters.
1. The page does not match visitor intent
Compare the search result, advertisement, referral or social promise with the landing-page heading and first paragraph. A service-specific visitor should not have to decode a broad homepage before finding relevance.
2. The value proposition is too broad
Explain who the service is for, the problem it addresses, what changes and why the approach is credible. Replace untestable phrases such as “innovative solutions” with specific scope, process or evidence.
3. One page tries to serve every audience
Different buyers bring different questions and authority. Give priority audiences a clear route, with content that addresses their decision without duplicating near-identical pages merely to target keywords.
4. Proof is distant, generic or unsupported
Place evidence near the claim it supports. Use approved case studies, dated results with context, credentials, process detail, screenshots or testimonials that can be verified. A logo wall may show familiarity, but it does not explain what was delivered.
5. The next step asks for the wrong commitment
A person diagnosing a problem may need a comparison or checklist before a sales conversation. A high-intent visitor may need a direct form or call. Match the call to action to the page’s purpose and explain what happens next.
6. The form creates avoidable friction
Every field should improve qualification, routing or response. Use clear labels and errors, preserve entered information after validation, test mobile keyboards and confirm the submission visibly. State who will respond and how personal information will be used.
7. Mobile visitors receive a weaker journey
Test on a real phone, not only a resized desktop window. Check tap targets, sticky elements, headings, accordions, chat widgets, validation, upload controls and confirmation states. Review the page on mobile data as well as office Wi-Fi.
8. The site is slow at the moment of decision
Use field performance data where available and identify page-level causes. Large media, third-party scripts and delayed interaction can interrupt the journey. Core Web Vitals are useful evidence, but pair them with conversion and user observations.
9. The page has no decision hierarchy
A visitor should be able to scan the problem, solution, proof, process and next step. Competing buttons, inconsistent headings and dense sections create extra decisions. Prioritise one primary action and make supporting navigation clearly secondary.
10. Measurement stops at page views
Track meaningful actions such as relevant service-page progression, form start, successful submission, booked meeting or call—then connect them to qualified outcomes where practical. Google’s GA4 guidance notes that a specific successful form event may be needed rather than treating every form submission as the same key event.
Leak diagnostic table
| Pattern | Evidence | First test |
|---|---|---|
| High exits immediately | Source-to-page message and first-screen content | Repair intent continuity |
| Engagement but no form start | CTA visibility, objections and proof | Clarify decision and next step |
| Form starts but no completion | Field-level errors and device testing | Remove or repair confirmed friction |
| Submissions but few sales contacts | Delivery, spam, routing and response logs | Repair lead operations |
| Many leads but weak qualification | Source, landing page and lead reasons | Improve intent and qualification |
Run the diagnosis in this order
- Verify the lead action and downstream delivery.
- Segment performance by source, landing page, device and lead quality.
- Watch representative users or run moderated task tests.
- Prioritise the leak with the strongest evidence and commercial effect.
- Change one material factor and define the decision rule before launch.
What the data cannot tell you alone
Analytics can show where behaviour changed, not always why. Low volume may make percentage movements unstable, and consent or tracking gaps can hide journeys. Pair quantitative data with sales feedback, user research and technical testing, and report uncertainty explicitly.
Official references
Use Website Redesign vs Conversion Optimisation to choose the intervention after diagnosis, or explore FGX’s web development services.