Lou Rosenfeld: «Redesign must die» – Summary from UXLondon

So Lou Rosenfeld told us that he was a hater of redesigns, and teach us easy methods to make a lot of small changes that make a lot of differences, instead of throw everything over board.

This is my attempt to summarize what Rosenfeld said, the «I» and «me» in the following text referring to Rosenfeld, not me (Ida)!

Wondering who I am? I’m an interaction designer working at NetLife Research. I’m @idaAaon Twitter, I also have a Twitter-account with links in English, called @idaAaLinks.

«I’m wearing two hats. One as a publisher, being happy, and a sad information architect. I need an information therapist. I’m kind of sad because I’m been dealing with a problem for 17 years now, called redesign.

I’ve been called a hater of redesign, but let me give you a little context of what I mean with redesign. I’m a consultant, so I work with huge, messy, unmanageable websites of huge organisations. Am I the only one, or has everyone run away and started doing app development? (Nope, I guess, since people are laughing, they’re not get rich on apps..)


I read about 12 years ago that the university of Michigan was going to redesign its site. They told the newspaper that they had hired my company for the redesign. They had a site that looked OK, but they had subsites that looked completely different. The complaints they were hearing were that people couldn’t find that they were looking for.

But I never heard from them. Basically they had some student labor and some software, so they had figured they didn’t need our help after all. The actually had a ribbon cutting.

Then they did it all over again. And again. And again. The site is still getting complaints about people not finding what they needed.

Why do people redesign?

The IBM redesigns were probably coming from the CEO.

It’s not just a case of vanity, but insanity: I need this change tomorrow. And they do it again and again and again.
– Seagull managers, mutters Cooper with a laugh, sitting next to us.

A lot of us make a lot of money doing these redesigns, even though we know it’s not they right way to go. So what are we doing? So many times we try to «boil the water», throwing every thing aboard and starting a new. This is attempting the impossible.
We also belive the unbelievable. There are often «solutions» from software vendors, but they are rarely the solutions of the actual problem of the institution.

We are also being irresponsible. We often let thing go at the expense of end users. We serve ourselves.
It’s about high time that we do something about this.

Ban the word «redesign»

One thing is starting to see the problem differently. I want to ban the word «redesign». It has nothing to do with what the actual solution. What are we solving for? What are we going to change? Not change for change’s own sake.

Rosenfeld wants a new term. Our websites are things that we need to see not as singular, perfectable, things that can even be understood in a snapshot, they are shifting under your feet. They are complex adaptive systems.

There are so many factors, how can you perfect it, how can you say that it’s done? Because in 2 years, you have to do it again. Even if you have great content, and you leave it alone, that content will not be good enough in a few years. It won’t be acceptable.

The swing of the pendulum

Regime A is that we have to get control of the mess, but hat doesn’ work. Then we gets Regime B, we have to let the thousand  flowers bloom. And then people feel like they need to get control. It’s a pendulum movement, and a healthy organisation has a little bit of both, both centralisation and decentralisation.

Redesigns also piss people off. You’re going to lose a lot of friends, for instance with Gawker blogs like Lifehacker, which had a drastic drop in visits.

My point is that redesigns is a dangerous course. I want to argue for refinement in stead.


You need to prioritize. Identify the important problems regularly A little goes a long way, think about Pareto principle and the Zipf distribution. A few queries amount for a huge amount of traffic. So we should love the short head, and leave the long tail alone. (This is similar to Gerry McGoverns argument about the «long neck» of websites).This way you can start improving the first 1% or 10% og 30% of the queries on your website. (While Ove Dalen at NetLife Research would suggest you spend a lot of time deleting the remaining 80%, like they did at Telenor).

The Zipf distribution is every way, a few search queries go a long way, a few documents, a little navigation. We also need to look at what are the important audiences, ans the essential wants and needs they have, and then combine and prioritize these wants and needs. Doing this we need information from both user research and stakeholder research. This can also be combined with user testing to see how big of a problem each of these are (how horrible is the current solution?)


You have to constantly improve your site, and address those problems regularly.Be an incrementalist. Repair the plane while you’re flying.

With the example from the University, they also found that the students had different needs through the year, and the data from the site shows the seasonality of their use. You have behaviors you need to tune for. We should also tune our processes. I committed a great crime when I co-authored the a book about IA. These days I’m advocating that you look at a content inventory as a process, that you’re constantly looking at.


Be an opportunist: find the low-hanging fruit.

Looking at wants and needs.. small changes that have big impacts. If you know that 4000 people a month are trying to find information about grades, that doesn’t sound like a lot, but it adds up. Another example is to teach the search engine to be smart, for instance like they have done at HP. If you put in a spell checker in the search for instance..

Wrap up! What shoud this mean for us?

First of all, design gets a lot easier. We’re coming up with small fixes, tweets. We’re making design challenges easier, but we have to handpick our methods.

Then there’s the hardest thing, which is the ethical side. We are ethically challenged. If you’re at an agency, you can earn a lot on redesigns. If you’re on a team, it’s the same challenge. It might be comfortable, but you’re not serving neither your employer nor your users. We’re getting to a point were organisations might realize that they don’t need to do redesigns.

I’m not worried, though. I have a lot of faith in the kind of people who are in this field. A lot of us are gap fillers. You probably got here because you find something unsettling. We are foxes, not hedgehogs. We know many little things. If we truly care about huge websites, then we’re going to take another approach.»

Wondering who I am? I’m an interaction designer working at NetLife Research. I’m @idaAaon Twitter, I also have a Twitter-account with links in English, called @idaAaLinks.

2 kommentarer om “Lou Rosenfeld: «Redesign must die» – Summary from UXLondon

    1. I guess they are Rosenfelds sentiments though, these are quite confusing notes I guess, since it’s actually more of a transcipt than notes.. The «I» and «me» in this text refers to Rosenfeld :)

Legg igjen en kommentar til Ida Aalen Avbryt svar

Fyll inn i feltene under, eller klikk på et ikon for å logge inn:


Du kommenterer med bruk av din WordPress.com konto. Logg ut /  Endre )


Du kommenterer med bruk av din Google konto. Logg ut /  Endre )


Du kommenterer med bruk av din Twitter konto. Logg ut /  Endre )


Du kommenterer med bruk av din Facebook konto. Logg ut /  Endre )

Kobler til %s