Most website decisions start with the wrong evidence: the design feels old, the owner is tired of looking at it, or a redesign would appear more professional.
Those may be reasons to pay attention. They are weak reasons to choose the size of the project.
Fix isolated failures. Rebuild when structural failures repeat across the site. Use a landing page when one campaign, offer, audience, or referral path needs a focused destination.
Old is not the same as broken. New is not the same as usable.
Judge the pattern, not the site's age#
The useful question is not whether the site has problems. Every site does. The question is whether those problems share a root.
Open the site and inspect this buyer path:
- The homepage opening screen.
- Navigation and page structure.
- The most important service page.
- Proof placement.
- The contact, booking, or inquiry path.
At each point, write down the first meaningful failure. Do not record every questionable font size and slightly awkward sentence. Record the first thing that would make a good-fit buyer hesitate, get lost, or fail to act.
Then compare the notes.
Different, contained failures usually support targeted fixes. A vague homepage opening and one broken form can be repaired without replacing a navigation system that still makes sense.
The same fit, structure, proof, or routing problem appearing throughout the path supports a rebuild. If every page struggles to explain the current business and send the buyer somewhere sensible, separate patches will keep treating the symptom.
A failure limited to one campaign or offer points to a landing page. The rest of the website does not need to be reorganized to give one audience a direct route.
There is no magic failed-page count. Judge the shared cause, not the tally.
Choose Website Fixes when the foundation still works#
Use targeted Website Fixes when the business and the basic site structure still match, but something specific is failing.
That often looks like this:
- One form, booking link, or phone action is broken.
- The homepage opening is unclear, but navigation and deeper pages remain coherent.
- One important service page is weak.
- Proof exists but is poorly positioned.
- Tracking, mobile behavior, speed, or another technical failure is isolated.
- The current business, service mix, and site structure still broadly match.
- Fixing the problem improves the existing system instead of adding another workaround.
A broken form is not a reason to rebuild a website. Neither is one vague headline. Fix the failure you actually have.
If the uncertainty is limited to the first screen, run the Homepage Clarity Test. That diagnostic owns the opening-screen repair. This decision is simply whether the problem stops there.
The key word is contained. A repair should make the existing site work better without forcing strange new menus, duplicate pages, or exceptions that future-you will have to remember.
Choose a Website Build when the structure is the problem#
Rebuild when the site is not merely presenting the business badly. Rebuild when it is presenting the wrong structure for the business.
The structural signals are usually hard to miss once you stop judging the paint:
- The business has materially changed, but the site still reflects the old version.
- Navigation no longer matches how the services are sold.
- Important service or decision pages are missing.
- The same fit, process, proof, and next-step confusion appears across several core pages.
- Proof is fragmented because the site has no clear places for it.
- Contact and booking paths vary unpredictably.
- Each proposed repair requires another workaround.
- A sensible page plan cannot be implemented cleanly inside the current structure.
If pages are missing or the current structure cannot support a sensible plan, use the page-map test to determine the smallest useful page structure.
This is a buyer-facing problem. The website makes people learn an outdated version of the business, hunt through categories that no longer fit, or guess which inquiry route is current.
I would not recommend rebuilding a site merely because it looks dated. I would recommend it when the structure keeps forcing the business to explain around it. If every sales conversation starts with “the website says that, but what we really do is...,” the site is no longer carrying its share of the explanation.
A rebuild should remove those repeated contradictions. It should not be a large visual refresh that preserves the same confusing routes underneath.
Choose a Landing Page when one path needs focus#
Use a landing page when one campaign, offer, audience, or referral path needs a dedicated destination and the current website can otherwise remain intact.
That page should make one promise, ask for one primary action, and avoid requiring changes to the rest of the site. A landing page is not a cheaper substitute for a missing website. It is a focused route for one specific decision.
If that is the lane, use the narrower guide to decide whether you need a new landing page or a better one. The detailed page decision belongs there.
The project-size decision at a glance#
| What is happening | Best starting scope | Why |
|---|---|---|
| One form, link, or technical path is broken | Website Fixes | The failure is isolated |
| The homepage opening is unclear | Website Fixes | Test and repair the first screen first |
| One service page is underperforming | Website Fixes | The rest of the site does not need replacement |
| Services changed but navigation and pages did not | Website Build | The structure represents the wrong business |
| Core pages repeat the same clarity and proof failures | Website Build | Separate patches would preserve the underlying problem |
| Important decision pages do not exist | Website Build | The site needs a new page system |
| One campaign or offer needs a focused destination | Landing Page | The project has one audience, promise, and action |
The table confirms the pattern test. It does not replace judgment about why the failure keeps happening.
Do not mistake design fatigue for a diagnosis#
Visual age can be a symptom. It is not the diagnosis.
None of these facts alone justifies a rebuild:
- The site looks dated.
- The owner is tired of the design.
- A competitor launched something prettier.
- One page is slow.
- One integration broke.
- The platform is unfashionable.
- A person selling websites recommended a rebuild.
Any one of them can prompt a closer look. None proves the buyer path needs a new structure. A patch should not become an excuse to preserve a weak structure, but a rebuild should not become a reward for being tired of the design either.
The opposite mistake is just as expensive. Do not preserve a bad site because it technically loads or because the business already spent money on it. Past cost does not make the current structure useful.
Run one final project-size test#
Open the homepage, navigation, most important service page, proof path, and contact path. Write down the first real failure in each.
If the failures are contained, fix them. If they keep pointing back to the same missing structure, stop stacking repairs and scope the rebuild.