A website redesign is a content project

We publish eight to twelve weeks for a website project. That number is honest about the part we control: design and build, done by people whose only job that month is your site. What it cannot cover is the part that decides most timelines, which is whether the words exist.

So the real answer to how long a redesign takes is a question about your side of it. Not how many pages, and not how complicated the design is. How many pages need writing, who is writing them, and what else that person has on in the same six weeks.

How long does a B2B website redesign take

Eight to twelve weeks is a fair working estimate for a mid-sized B2B site, and it holds when content arrives roughly on schedule. When it does not, very little downstream can proceed, because almost every remaining task takes real copy as an input. Templates cannot be proved, page lengths cannot be settled, and nothing can be signed off in a way that survives contact with the actual text.

The asymmetry underneath this is worth naming. Our work is done by people booked to do it, so it is predictable. Your content is produced by people who already have a full week without you, so it is not. A marketing lead who owes nine pages is not being obstructive when those pages take a month. They are doing their job and this one, and only one of the two has somebody chasing it.

It is why "the site is nearly done" can be true and useless at the same time. The design is finished, the build is finished, and eleven pages are waiting on a paragraph about what the product actually does.

Why the copy is the thing that slips

Rarely because anyone is a slow writer.

Ownership is the first cause. Content tends to belong to a team rather than a person, which means it belongs to nobody with a diary. We have argued elsewhere that a rebrand needs one decider, and the same logic applies to the words, at a smaller scale and with more immediate consequences. A late brand decision is expensive. A late paragraph stops the project.

The second is that writing the site is not a formatting task. Working out what goes on the product page means settling what the product is for and who it is not for, and those questions were usually unresolved long before anyone mentioned a redesign. The project did not create the disagreement. It put a date on it.

The third is that the tools let you defer it. A layout filled with placeholder text looks finished. It presents beautifully in a meeting and it conceals every problem the real words would have exposed: the heading that is four words in the design and nineteen in life, the three-column section with two case studies to put in it, the feature list that turns out to be a feature.

A design approved on placeholder text has not been approved. It has been admired.

That is why our website process puts real content into the design stage rather than after it, and why structure comes before surface. It is not a purist position. A template proved with real words is a decision. A template proved with Latin is a mood board.

Most B2B website problems are content problems

Nielsen Norman Group's B2B website usability report, now in its third edition, is built on testing 293 business sites across the United States, the United Kingdom and Singapore. Look at what it covers: technical specifications, product comparison charts, case studies, how pricing is displayed, what a site needs to say at each stage of a long buying cycle. Very little of that is styling. It is information, and whether a stranger can get to it.

Buyers are not deciding whether your site is attractive. They are working out whether you can do the specific thing they need, and specificity lives in sentences. A redesign that improves the typography and reprints the same vague claims has changed the packaging on the problem.

How to run a redesign that finishes

Do the inventory before you commission anything. List every page the new site will have and put a name and a date beside each one. A name, not a department. The list is normally longer than anyone expected, and that is a discovery worth making in week zero rather than week six.

Write the page you have been avoiding first. Most companies have one, usually the page explaining what they do or who it suits. If it is hard to write, the difficulty is the point: something is unresolved, and it will not resolve itself once forty other things depend on it.

Then cut. A twenty page sitemap with nine honest pages and eleven placeholders is worse than a nine page site, and nobody is obliged to fill a navigation. Pages can be added afterwards by a team trained to add them, which is the normal state of a site being looked after rather than rebuilt every three years.

Our client suitability page says we cannot do three months of thinking in a fortnight and keep a straight face about it. This is the other half of that sentence. Neither can you, and a redesign is the thing that tends to find out how much thinking you had left to do.