A website that is not generating leads does not automatically need a redesign.
First find out where the lead journey is failing.
There are three very different situations that business owners often describe with the same sentence: “my website is not working.”
- The right people are not reaching the site.
- The right people arrive but do not enquire.
- People try to enquire, but the form, tracking or follow-up system fails.
A new visual design only directly addresses part of the second problem. If you redesign before diagnosing the failure, you can spend money and preserve the real problem underneath a newer interface.
Start by separating traffic problems from conversion problems
Open your analytics and search data before changing the homepage.
If the website receives almost no relevant visits, low lead volume may be a discovery problem. The next investment may be stronger service pages, local visibility, paid acquisition or search content rather than a redesign.
If relevant commercial pages already receive qualified visits but almost nobody calls, messages or submits a form, you have a conversion problem.
If analytics says nobody converts but enquiries are arriving in your inbox or WhatsApp, you may have a measurement problem.
That distinction matters because each problem needs different work.
A useful first audit is:
| What you see | Investigate first |
|---|---|
| Almost no relevant traffic | Search visibility, targeting, distribution and service-page coverage |
| Traffic but few enquiries | Offer clarity, proof, CTA, friction and intent match |
| Form starts but few completions | Form length, validation, mobile usability and errors |
| Customers say they submitted but you received nothing | Form delivery, email routing, backend logs and spam handling |
| Leads arrive but analytics shows zero | Conversion-event implementation and thank-you/success state |
| Enquiries arrive but are poor quality | Traffic intent, positioning, qualification and offer clarity |
Do this before debating fonts, colours or animations.
Test whether the enquiry mechanism actually works
This is the least glamorous check and one of the most important.
Submit every important form yourself.
Do it on a phone and desktop. Use realistic data. Trigger validation errors deliberately. Confirm the success message appears. Then confirm the enquiry reaches the actual inbox, CRM or database a human monitors.
Google’s web.dev form testing guidance recommends testing forms across devices, input methods, browsers and operating systems, then measuring their performance. Its accessibility guidance also emphasizes clear errors and a success state after a valid submission.
A form that looks correct but silently fails is worse than an ugly form that works.
Also test alternative lead paths:
- click-to-call links;
- WhatsApp buttons;
- booking links;
- email links;
- quote-request buttons;
- calendar integrations.
Do not count a button click as a successful lead when the actual business outcome happens several steps later.
Make sure you can measure a real lead
You cannot improve a lead journey confidently if you only measure page views.
Google Analytics 4 recommends the generate_lead event when a user submits a form or request for information. Google’s recommended events documentation also supports later lead stages such as qualification and conversion when a business needs a fuller funnel.
For a simple service website, you should at least be able to answer:
- Which page did the visitor enter through?
- Which service or offer were they viewing?
- Did they reach the enquiry mechanism?
- Did the submission actually succeed?
- Which channel brought them there?
If you cannot answer those questions, install or repair measurement before making a large redesign decision.
Otherwise, “the redesign worked” can become a feeling rather than an observable result.
Check whether the page answers the buyer’s first questions
A visitor should not have to decode the business.
On an important service page, ask:
- Can I tell what this business does quickly?
- Can I tell whether it serves someone like me?
- Can I understand the outcome, not only the service name?
- Is there evidence that the business can deliver it?
- Do I know what happens if I enquire?
- Is the next action obvious?
Many conversion problems are messaging problems wearing a design costume.
A polished hero that says “Empowering Tomorrow Through Innovative Solutions” may look modern while telling a buyer almost nothing. A more specific headline can outperform a prettier but vague one because it reduces uncertainty.
This is why I would usually fix positioning and page hierarchy before replacing the entire technology stack.
Inspect the intent of the traffic, not just the traffic number
A thousand visits can produce fewer leads than fifty high-intent visits.
Suppose an accounting firm ranks for a broad educational article that students frequently read. Traffic rises. Enquiries do not. That does not prove the website has a conversion problem.
Now suppose the firm’s “outsourced bookkeeping for small businesses” service page receives visitors searching for providers, but almost nobody enquires. That is much stronger evidence of a conversion issue.
Separate informational pages from commercial pages. Then inspect the search queries, ad terms, referring sources and geography behind the visits.
This principle matters for SEO too: more impressions are useful only when they help the business reach the right people or build authority that supports that acquisition system.
Look for trust gaps near the decision point
People hesitate when the website asks them to take a meaningful action without reducing enough uncertainty.
The right proof depends on the business.
A consultant may need clear expertise, named experience and case outcomes. A contractor may need real project photos, service areas and credentials. A clinic may need practitioner information, location details and a clear appointment process. A software builder may need credible work, process clarity and evidence that ownership and handoff have been considered.
Do not add decorative testimonial sliders simply because “social proof” appears on a checklist.
Ask what a cautious buyer would need to verify before contacting this particular business.
Then put that proof close to the decision it supports.
Audit mobile as a customer journey, not a screenshot
A website can be responsive and still be difficult to use on a phone.
Try to complete the actual task one-handed:
- understand the offer;
- compare the relevant service;
- inspect proof;
- open the form;
- type into every field;
- recover from an error;
- submit successfully;
- call or message through an alternative route.
Pay attention to sticky elements covering controls, keyboards hiding fields, tiny tap targets, layout shifts, slow media and forms that require unnecessary typing.
Google’s PageSpeed Insights reports both lab and real-world performance data, while Google’s Search documentation recommends good Core Web Vitals as part of a broader page experience rather than treating a performance score as the whole answer.
Speed matters, but a 100 performance score will not rescue an unclear offer.
When should you fix the existing website?
Choose targeted fixes when the underlying site is basically healthy and you can identify a limited set of failures.
Examples:
- the contact form does not deliver reliably;
- the main CTA is buried;
- service copy is vague;
- mobile spacing makes the form frustrating;
- important proof is missing;
- analytics does not record successful enquiries;
- one or two high-intent landing pages do not match their traffic;
- an oversized hero image or third-party script is causing avoidable performance problems.
In these cases, rebuilding everything introduces unnecessary cost and regression risk.
Fix the broken boundary, measure it, and learn from the result.
When is a redesign justified?
A redesign becomes more sensible when the problem is structural across the customer journey.
For example:
- the navigation reflects internal company departments rather than customer needs;
- several services are difficult to distinguish;
- the site has no useful hierarchy between discovery, proof and action;
- the visual system consistently makes important information hard to scan;
- mobile layouts require page-by-page workarounds;
- the website was built for a different business model and the current offer no longer fits it;
- campaigns repeatedly need separate pages because the core site cannot support a coherent buying journey.
A redesign should therefore begin with information architecture and conversion logic, not a mood board.
The question is not “how can this site look newer?”
It is “how should a prospective customer move from need to confidence to action?”
When should you rebuild instead of redesign?
A redesign changes the experience. A rebuild changes the underlying implementation.
You may need both, but do not confuse them.
A rebuild is easier to justify when the existing technology prevents reliable improvement, for example when:
- nobody controls or can safely deploy the source code;
- the CMS makes normal business updates impractical;
- basic fixes repeatedly break unrelated pages;
- the platform cannot support a required booking, payment or integration flow safely;
- security or dependency problems cannot be maintained reasonably;
- performance is constrained by architecture rather than a few fixable assets;
- ownership, hosting or deployment is tied to an arrangement the business cannot control.
If the existing stack can support the desired customer journey reliably, rebuilding it simply because another technology is fashionable is usually wasteful.
If you are considering moving platforms, use my website migration checklist before changing URLs or infrastructure. A rebuild should not casually destroy search equity, analytics continuity or working customer journeys.
A practical order of operations
If your website is not generating leads, I would work in this order:
1. Verify measurement
Confirm visits, successful form submissions, calls or other meaningful lead actions are actually tracked.
2. Verify delivery
Make sure the business receives the enquiry. Test the complete path, not just the button.
3. Verify traffic intent
Check whether the people reaching commercial pages could realistically become customers.
4. Fix clarity and proof
Make the offer, audience, outcome, evidence and next action easier to understand.
5. Remove interaction friction
Test mobile, forms, speed, validation, navigation and accessibility.
6. Improve the highest-value pages first
Do not redesign twenty pages when two service pages create most of the opportunity.
7. Redesign only when the structure itself is the problem
If fixes keep exposing the same cross-site hierarchy and journey problems, redesign with those findings as requirements.
8. Rebuild only when the implementation blocks the solution
Technology should follow the business and product requirement, not lead it.
Do not pay for a redesign without defining success
Before a redesign starts, write down what should be different afterwards.
Not “the site should feel premium.”
Use outcomes such as:
- more qualified quote requests from service pages;
- fewer failed form submissions;
- a clearer mobile booking path;
- measurable calls and WhatsApp enquiries;
- better conversion from a specific campaign landing page;
- easier publishing by the internal team;
- improved performance without breaking important functionality.
Then preserve a baseline so the new version can be compared against it.
That turns web design from a cosmetic project into a business intervention.
If you are deciding whether the current site needs targeted repair, a redesign or a rebuild, my web design service is built around that diagnosis rather than assuming every project needs a new site. You can also contact me with the current URL and the business outcome you are trying to improve.
