A website redesign should start by recording what currently works — your best-performing pages, your existing URLs and your current traffic — so none of it is lost in the rebuild. The most common and most damaging mistake is launching without mapping old URLs to new ones. A&B Consulting Group plans and builds redesigns from Burr Ridge, Illinois, serving Chicago and the western suburbs.
Most redesigns are judged on how the site looks and that is why so many of them lose traffic. The new site is better in every way anybody discussed, and three weeks later enquiries are down by half, because the pages that were quietly producing them no longer exist at the addresses Google knew.
This is the checklist we work through. The order matters more than the contents — almost everything that goes wrong in a redesign goes wrong because something in the first phase was skipped.
Launching without a redirect map. It is unglamorous work and it is the single highest-stakes task in the project.
Every page on your current site has an address Google has indexed, other sites link to, and customers have bookmarked. Change the addresses without telling anybody where things went, and every one of those becomes a dead end. The rankings do not transfer, the links stop passing authority, and the traffic goes to whoever is still reachable.
The damage is also slow to appear, which is why it so often goes undiagnosed. Rankings decay over weeks, so the decline gets blamed on seasonality or the market long after the cause is fixable cheaply.
A redesign is the expensive answer and is frequently aimed at the wrong problem.
If the site looks dated but converts well, the honest move is usually to leave it and spend the money on getting more people to it. If nobody is finding you, that is a search and content problem, and a new design does not create pages that do not exist. If enquiries arrive and sit unanswered until Thursday, nothing about the website changes that.
The cases where redesign genuinely is the answer: the site is slow enough to lose people, it is not usable on a phone, it has no structure for the services and locations you need to rank for, or it cannot be edited without a developer. Those are structural and cannot be fixed by spending more upstream.
Redesigns stall for organisational reasons more than technical ones, and the two predictable failures are worth naming up front.
Content: someone internally has to own supplying or approving copy, with a date. In most late projects the delay is a page of text waiting on a person who has a day job.
Decisions: one person with authority to approve. Design by committee produces a site that averages five opinions and expresses none, and it adds weeks.
From our side you get one point of contact rather than a ticket queue, and we would rather tell you a deadline is slipping because copy has not arrived than let it slide quietly.
It can, and the cause is almost always the same: URLs changed without redirects mapping old addresses to new pages. Done properly — URL structure preserved where it works, ranking content carried over, every old URL permanently redirected to its closest equivalent — a redesign should hold rankings and usually improves them through better speed and structure.
A few weeks for a standard custom build once content is settled. Content is almost always the bottleneck rather than design or development, which is why we agree what the pages are and who writes them before anything gets built.
Where they work, yes. Changing a URL purely for tidiness costs ranking and gains nothing a visitor will notice. Change them when the structure is genuinely wrong — no service or location pages, for instance — and redirect every old address to its closest new equivalent when you do.
A list pairing every old URL with the single most relevant new page. You need one for any redesign where addresses change. Redirecting everything to the homepage instead is treated as a soft 404 and throws away the authority those pages had earned.
By comparing against the figures you recorded before starting — traffic and enquiries page by page rather than in total, plus rankings for your main terms. Without that baseline you are guessing, which is why phase one of the checklist is recording rather than planning.
Improve it, unless the problems are structural. Slow, unusable on a phone, no structure for the pages you need to rank for, or uneditable without a developer — those justify a rebuild. Looking dated while converting well does not, and that money is usually better spent getting more people to the site.
Tell us what you have now and what you are trying to fix. We will scope it honestly — and if you do not need us yet, we will say so.