Website redesign vs incremental improvement: a decision framework for established organizations

The wrong reason to redesign

“It looks dated” is the most common reason given for a redesign and the least useful. Sometimes it is true and it matters, because buyers judge a firm by its site and a tired site makes an established business look smaller than it is. But it is also the reason most likely to produce a new site with the same problems in a fresh coat of paint: the same unclear service pages, the same generic contact button, the same lack of measurement.

The opposite mistake is just as common. An organization keeps patching a site that is fundamentally broken, adding plugins and landing pages to a platform nobody fully understands, because a redesign feels too big to start. Every fix makes the next one harder.

The way out of both is to diagnose before deciding. Look at the site through four lenses, and be honest about which problems are structural and which are specific.

Lens one: technical debt

Technical debt is the accumulated cost of past shortcuts: outdated platforms, abandoned plugins, custom code nobody documented, hosting nobody remembers setting up. It is the lens most likely to force a redesign, because some of it cannot be fixed in place.

Signals that point to a rebuild:

• The platform or theme is no longer supported, or core updates are being avoided because they break things.
• Only one person (often a former contractor) knows how the site works.
• Routine changes require a developer, and every change risks breaking something else.
• Performance problems are baked into the theme or page builder rather than caused by a few heavy images.
• You cannot meet an accessibility standard without rebuilding the templates.

Signals that it can be fixed incrementally: the platform is current and supported, the site is slow because of specific assets or scripts, a few plugins need replacing, or hosting needs upgrading. These are maintenance projects, not redesigns.

Lens two: conversion problems

If visitors arrive and do not act, the question is whether the problem lives in a few places or everywhere.

Specific conversion problems are good candidates for incremental work: a form with too many fields, a service page that buries the call to action, a mobile layout that hides the phone number, a landing page that does not match the ad that sent people there. Each can be fixed, measured and improved on its own. The conversion rate optimization playbook describes how to run that cycle, and what makes a B2B website convert covers the elements to look at first.

Structural conversion problems are different. If the navigation is organized around your internal departments rather than what buyers are looking for, if there is no path for a visitor who is not ready to talk yet, or if every page funnels to the same generic contact form, the issue is the architecture. You can tune individual pages for a long time without fixing a site that is organized around the wrong idea.

One honest caveat: if you do not have reliable tracking, you cannot tell which kind of problem you have. Fixing measurement is almost always step one, whichever way you go. If conversions have recently dropped rather than always been weak, start with how to diagnose a drop in website conversions before assuming the site is the cause.

Lens three: content

Content problems are often mistaken for design problems. A site full of vague copy, outdated case studies and service descriptions written for insiders will underperform in any template.

Incremental improvement suits content that is mostly sound: rewrite the weakest service pages, add proof where it is missing, retire stale news, tighten headlines. This work can start next week and does not require a new site.

A redesign becomes more likely when the content problem is systemic: the site has grown to hundreds of pages that nobody owns, there are several pages competing for the same topic, the structure cannot accommodate what you now sell, or content needs to exist in a second language and the current platform cannot handle it properly. At that point, rewriting page by page inside a broken structure wastes the writing.

Lens four: business goals

This is the lens that most often justifies a full redesign, and the one owners are best placed to judge. Has the business changed in a way the site does not reflect?

• A rebrand or repositioning that changes what you say and who you say it to.
• A new service line, a merger or an acquisition that changes the structure of the offer.
• A move upmarket, where the site needs to persuade a more senior or more sceptical buyer.
• A new market, such as selling into Quebec or recruiting nationally.
• A shift in how sales happen, for example from relationship referrals to inbound leads.

If the answer is yes, incremental changes will tend to produce a site that says two things at once. If the business is broadly the same as when the site was built, and the site simply needs to do its job better, improvement is usually the better investment. If a rebrand is on the table, settle it first; see when to rebrand an established business.

A simple scoring method

Put the four lenses in a short table on paper and mark each one: no real problem, specific problems, or structural problem. Then read the pattern.

• Structural in technical debt or business goals: redesign, or at least re-platform. These are the two lenses that incremental work cannot fix.
• Structural in conversion or content only: consider a structural rework (new architecture and templates) on the existing platform if it is sound. This is the middle option people forget exists.
• Specific problems across the board: improve incrementally, in a prioritized backlog, with measurement in place.
• No real problems, just fatigue: refresh the visual layer and spend the rest on content and promotion.

An illustrative example: a regional professional services firm with a site about seven years old. The platform is current and well maintained, so technical debt is low. Conversion is weak, but tracking shows the issue is concentrated on three service pages with no proof and a generic form. Content is thin in places. The business has not changed. The verdict is incremental: rewrite the key service pages, add case studies, split the form by inquiry type and measure for a quarter. A full redesign would have cost far more and fixed the same three pages.

Now change one fact: the firm has merged with a practice in another province and now offers services it did not before. The business-goals lens flips to structural, and a redesign is the right call, because the architecture needs to carry an offer that did not exist when it was built.

The hybrid most organizations need

In practice the best answer is usually both, in sequence. Fix the foundation once, properly, and then stop treating the website as a project that happens every five years.

A redesign done this way is smaller than people expect. It focuses on what genuinely needs rebuilding (platform, architecture, templates, measurement) and leaves room in the budget for the improvement work that follows. After launch, a steady rhythm of changes, each one measured, keeps the site from sliding back into the condition that triggered the redesign. That is how we approach website design and development with established clients: build the foundation to last, then improve it on evidence.

What continuous improvement looks like in practice

It is less glamorous than a launch, which is partly why it gets skipped. It needs three things:

• A backlog. A running, prioritized list of changes, each with the problem it addresses and how you will know it worked. Ideas from sales, front-line staff and analytics all go on it.
• A cadence. A monthly or quarterly slot where a few items get done, rather than waiting until enough frustration builds up to justify a project.
• An owner. One person inside the organization who decides what goes next. Without that, the backlog becomes a wish list and the site drifts.

Organizations that run this rhythm tend to find that their next “redesign” is much smaller, or never arrives at all.

If you do decide on a redesign, the website redesign planning guide covers what to settle before design begins.

What to do first

Before you talk to any vendor, spend an hour on three things. Check whether your platform and theme are current and supported. Check whether your analytics can tell you how many qualified inquiries the site produced last quarter. And write two sentences on whether the business has changed since the site was built. Those three answers will tell you which direction to lean, and they will make every vendor conversation that follows more useful.