A redesign wraps up the way most projects do. The new site goes live, the team reviews it together, everyone agrees it looks better than before, and the project gets marked complete.
That moment of approval is real. But it measures something narrower than most people realize.
A redesign passes the internal test when everyone approves it. It passes the real-world test only after strangers try to use it.
Two Different Tests
The redesign test is the one everyone runs, often without naming it as a test at all. Does it look better? Was it delivered on time? Do stakeholders approve of the new design? Is the site live and functioning?
Those are reasonable questions, and a good redesign usually answers all of them well.
The real-world test is different, and it doesn’t happen in a meeting. Can a visitor who has never seen the business before understand what it does within a few seconds? Can they find what they came for without hunting? Do the contact paths actually work? Does the information stay accurate once the launch excitement fades and nobody is actively watching the site anymore?
The first test gets run deliberately, on a schedule, by people who already know the business. The second test runs constantly, informally, by strangers, and nobody is scoring it.
Why the Gap Exists
This isn’t a failure of the people who built the site. A redesign project has a defined scope and a defined end point. Someone approves the final version, the invoice gets paid, and the relationship moves on to the next thing. That structure measures completion, not ongoing usefulness.
Approval also happens inside the business, by people who already understand the products, the services, and the intent behind every page. That’s a fundamentally different vantage point than a first-time visitor’s. The internal team is well positioned to judge whether the site looks professional and reflects the brand accurately. They are poorly positioned to notice whether a stranger can find the contact page in under ten seconds, because they never have to try.
A redesign can be delivered professionally and still leave real gaps, simply because the handoff was structured to measure completion, not the ongoing experience of people encountering the site for the first time.
What the Redesign Test Misses
A new site can look clean and modern while the actual copy still assumes the reader already knows what the business does. A navigation menu can be visually consistent with the new design system while still organizing pages by internal department rather than by what a visitor is trying to accomplish. A contact form can be rebuilt with better styling while the underlying issue, whether submissions reach someone who reads them, was never part of the redesign scope at all.
A new site can also inherit old content, old assumptions, and old gaps under a cleaner surface.
None of this shows up in a stakeholder review, because a stakeholder review is checking against a different list.
The same pattern shows up whenever a project has a clear finish line: not just full redesigns, but launches, platform migrations, and content refreshes. The specific project changes, but the underlying gap is the same. The team measures what was delivered. Nobody measures what happens after.
Why It Surfaces Months Later
In the weeks right after launch, everyone is still paying close attention. Traffic gets checked, feedback gets requested, small issues get caught and fixed quickly. That attention fades on its own timeline, usually faster than anyone expects.
Once that attention moves on, the site keeps running exactly as built. If a visitor gets confused by the navigation or can’t find a way to make contact, there’s no meeting where that gets raised. The site looks the same as it did on launch day, still reflecting the version everyone approved. What’s changed is that nobody is checking it against the original questions the redesign never fully answered.
Running the Real-World Test
The most direct way to close this gap is to test the site the way a stranger would, separately from whether it looks right internally.
That means reading the homepage as someone with no prior context would, and checking whether it’s clear within a few seconds what the business does and who it’s for. It means trying to complete the task a typical visitor would come to do: find a service, check pricing, get in touch, and noting where that path gets confusing or slow. It means testing the contact form and the phone number the same way, actually submitting and calling, not just confirming they display correctly.
None of this requires rebuilding anything. It requires running a different test than the one the project was scoped around, and running it periodically rather than once at launch.
The Project Ends. The Website Doesn’t.
A redesign has a finish line because it needs one to get scoped, budgeted, and delivered. That’s a reasonable way to run a project.
But a website doesn’t stop being used once the project closes. It keeps being encountered by people who have never seen it before, on a schedule nobody controls. The redesign test only has to be passed once. The real-world test runs every time someone new arrives, and it’s worth checking on more than once.