Are you about to launch yet another web redesign project? If so, think again.

One of the reasons organisational websites fester and decay is because companies are good at projects and poor at iteration.

Most organisations like to think in terms of clearly defined projects. In many ways this makes a lot of sense. Projects are easily quantifiable in terms of the budget, resources and time involved. It is easier to find budget and resources for finite projects rather than ongoing investment.

This is why redesign projects are so common within web design. Organisations love them because they have a clearly defined scope and provide an easily identified deliverable.

In short, for specific investment you can see a tangible change.

The web is an iterative environment

Unfortunately the web is not ideally suited for project-based development. Instead the nature of the web encourages iteration. This is because things evolve so quickly online that periodic redesigns are not enough to keep a website fresh and meeting users rapidly evolving expectations.

Not only that, but the web provides some unique tools that are ideally suited to iterative development. For example, the web allows you to measure changes with an accuracy unknown to other mediums.

Small changes are more measurable

The web is better suited to measuring small incremental changes than it is large redesigns. When you overhaul an entire user interface is impossible to identify which changes have generated the most significant return on investments.

However, making a small change allows the measurement of that specific item and what difference it makes to conversion.

For example, I once worked on an ecommerce site where changing the micro-copy relating to site security caused a 6% increase in conversion.

This small iteration told us how important micro-copy was on this particular site and we were able to make other improvements that also increased conversion. Had we done a complete redesign of the site this easy win would more than likely have been missed.

Rapid evolution rather than static projects

The message here is simple; we need to move away from treating the web as just another project to a more iterative process involving rapid evolution.

Our websites need to be a place where we iterate and test. Yes, this does mean an ongoing investment but does not need to be prohibitively expensive, especially when compared to periodic redesign.

There are many tools that make iteration and monitoring easy. Tools like or that help facilitate remote user testing or tools like that makes it easy to do split testing on any website.

These tools are inexpensive and can provide valuable data extremely quickly.

There is no excuse for only periodically redesigning your site. Why limit your website's optimal performance to immediately after a periodic redesign, when it could be running at peak performance the entire time?

Paul Boag

Published 18 April, 2013 by Paul Boag

Digital Strategist Paul Boag is the author of Digital Adaptation and a contributor to Econsultancy. You can follow him on Twitter or his personal blog Boagworld.

29 more posts from this author

You might be interested in

Comments (6)

Save or Cancel


I think it depends on the nature of the site and what you want to achieve with a redesign. If you're adding functionality, would it not be better doing it all in one go? As a user i would rather wholesale change to a site one morning than a changed feature everytime i visit.

over 5 years ago


Paul Boag

Hi Tom.
It is interesting that you say that as a user you would prefer to see a wholesale change rather than small incremental improvements.

This goes against the body of research that I have seen on the subject. People rarely respond well to large changes but often fail to notice minor updates.

In fact I would go as far as saying that this is yet another good reason for making incremental change.

You only need to look at the negative response which happens every time Facebook redesigns to see quite how negatively people react to the process.

over 5 years ago



Well i guess it depends on the nature of the updates - if it's a change in content formatting or hiding text behind a 'read more' button, then the chances are they wouldn't be noticed - but if it's a change in how you login, search, checkout etc then it probably would.

I think people are naturally adverse to change (hence the facebook compliants) but they get used to them pretty quickly, which is why i think it's bets to make wholesale changes in one go (kind of like ripping off a plaster).

A large part of it is how you manage change as well. You could offer a phased release with a beta site or the option to roll back to the previous version. You could explain the reasoning behind the change and give plenty of notice (the BBC did this very well). It's not a 'one size fits all' situation.

over 5 years ago



I think you are both right. I lean on Paul's viewpoint: redesigning an entire site is not a safe move, particularly if all of your business is done through your website (no brick-and-mortar store). This is a great viewpoint for "design" (colors/images), messaging, and page hierarchy. Basically, working with items that already exist on the site.

However, I've found with testing that when adding functionality or features (particularly ones that require databases, forms, script-writing, etc), it's sometimes easier to hardcode the change and then test. This way the element is done correctly so it works with everything on the site and exactly how you intended. I've found that testing technology is not built for creating big experience changes like this and you create more work trying to make the experience work within your testing technology's limitations rather than just hardcoding it.

In these instances, I find hardcoding first and then going back and removing the new feature as the test can be really effective. Plus, you are able to tweak the now-existing element better with different variations since it now exists on the site for you to implement your testing software code around.

Either way, you are still testing. You are hardcoding with the understanding that this could be removed at a later date when the test concludes.

I think this touches on the biggest disconnect in the CRO world: what does redesign mean? Often, it's assumed it means all kinds of rearranging, changing of elements, altering functionality, etc. As a professional CRO, my experience is that design means: colors, font types, images, etc. The hierarchy of the pages, layouts, prominent elements, etc are all testing realm, not design.

This definition is similar to an actual facelift. You are tweaking the look of something (removing wrinkles or whatever), but you aren't moving someones nose to their forehead and their lips on their chin...which is how I feel companies view "website redesigns."

over 5 years ago


Christopher L

One feature I have found gets lots of traffic is a "What's new on the site" link.

Otherwise I don't like it when a site redesigns, and I have to relearn navigational paths or search again to find things I previously knew how to locate.

So I am with Paul's point of view on this. Working with big brands in large agencies the work is project focused, and it's incredibly difficult to get budget for ongoing updates and maintenance. "When it's built it's done" is the attitude, and no client wants to allocate budget without knowing specifically what they will get for it.

Yet once a redesign or a new site is launched team members and users often suggest a flurry of improvements or tweaks that lack budget to implement, and more become obvious to do as time passes. Then it's a matter of bundling these together, and trying to get budget for a "new project". This is slow and inefficient, although it is probably easier to control budget.

over 5 years ago


Sarah A

I agree with Christopher, when you go for the project approach it is often difficult to get budget or enthusiasm for ongoing work. But I do also see that changing slowly with organisations looking to upgrade aspects of their site so that it better supports their social media for instance.

Somethings will always need to be projects I think ( a change of CMS, a major rebrand or merger, introducing or upgrading ecommerce) but the best approach I think is to have a rolling plan of marketing and digital marketing activities, so you can see how you are impacting on users and how well you are achieving your objectives.

over 5 years ago

Save or Cancel

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.