On not quite agreeing with Jakob Nielsen

Web usability guru Jakob Nielsen published an important post this week on the question of whether it's necessary to build more than one website to cope with today's mobile web. If you're a web designer or are commissioning a website you should read it: see Mobile Site vs Full Site.

Quite a few designers have indeed read it, and it's safe to say that it hasn't been universally popular, as selected tweets evidence. I could have picked out many more. Jakob Nielsen has always been a controversial figure within the web design world. His longstanding Alertbox column is deliberately provocative, often harshly critical of design sensibilities that don't strictly abide by his rather severe standards for website usability. Jakob, whose online persona evokes something of the air of a puritan Danish sermoniser, tends not to make suggestions: he makes rules.

Well thumbed

Inconveniently for designers desiring a free reign, many of Nielsen's rules are based on exhaustive user research undertaken by his agency, the Nielsen Norman Group. He certainly hasn't got everything right: notoriously, he persisted in arguing for many years that designers should never use link colours other than the default shade of blue, no matter whether it clashed horribly with the rest of the design. But many, many other recommendations that have emerged from the Group's research have been accepted and over time become core principles of website usability. I for one remember well the impact of Nielsen's book Designing Web Usability, published back in 2000. At that time it was widely recognised as the most important book every published about web design, and I think it should still be on any new designer's reading list.

So everything Nielsen says about the mobile web is well worth reading. And he has had a lot to say of late. The essential argument of this latest piece is that if you can afford it you should have separate websites optimised for the different strengths and weaknesses of mobile and desktop browsing devices. To summarise:

Good mobile user experience requires a different design than what's needed to satisfy desktop users. Two designs, two sites, and cross-linking to make it all work.

Much of the article is dedicated to outlining the significant ways in which the mobile web experience differs from that on the desktop, and to making the uncontroversial point that a website user interface must adapt to ensure its optimised for the device on which its being viewed. For example content must be as succinct as possible for all users, but especially for mobile users. Navigation patterns are different - there's a discussion of the 'fat finger' problem encountered by mobile users when faced with dense menus. And mobile users can't afford to download unnecessary features that might appear relatively unobtrusive on a larger desktop display. He writes:

The desktop user interface platform differs from the mobile user interface platform in many ways, including interaction techniques, how people read, context of use, and the plain number of things that can be grasped at a glance. This inequality is symmetric: mobile users need a different design than desktop users. But, just as much, desktop users need a different design than mobile users.

All very sensible, and based on a lot of user research. The controversy consists in Nielsen's concluding argument that it is therefore necessary to build separate sites for mobile and desktop:

[Y]ou could just optimize the entire website for mobile in the first place. Then, giving mobile users the 'full' site wouldn't cause them any trouble. While true, this analysis neglects the penalty imposed on desktop users when you give them a design that's suboptimal for bigger screens and better input devices.

The complaint is that here, and with his previous mobile web posts, Nielsen hasn't properly engaged with the responsive web design methodology developed by designers over the past couple of years, which has shown that in many cases - not all - it is possible to build a single site that adapts according to the device on which it is viewed. Responsive web design isn't a silver bullet, but it has most certainly emerged as a credible design solution for organisations that simply don't have the budget to develop and maintain separate websites. Google responsive web design or take a look at a showcase site like mediaqueri.es and you'll immediately see just how much work has and is being done to refine and further develop this approach.

Nielsen's rather peremptory dismissal of this important development in contemporary web design is, I think, bound up with his Group's focus on their core audience and client base: large corporations. The Group tends to recommend that clients take the most expensive, most conservative option, in this case the development of separate mobile and desktop sites: cost isn't really an issue. Responsive web design is, fundamentally, a methodology borne of necessity: a grassroots solution developed by designers working to tight budgets and schedules for clients who can only afford one website, and who need it to work as well as possible beyond the desktop.

I'll continue to subscribe to Jakob Nielsen's Alertbox, and greatly value the advice he and his Group have shared over the years. But unless you are working for a company with big pockets I wouldn't take all of it as gospel, no matter how authoritatively expressed.


Please leave a comment if you like. The comment field accepts basic HTML tags. You can also contact me at @justinlucent and Google+.



The first rule of Rule Club is: there are no rules. There are however many, many good practices (as demonstrated by mine host) . The second rule of Rule Club is: first know, what’s the name of the game (no, not the ABBA song).

Mobile, in the sense of the mobile web, is still in its infancy, and to me the biggest failing many ‘web types’ have made is to make the principle issue one of content (or perhaps partial content)  _display_.  We need to start thinking more about differences in responsive content _choice_. To do that efficiently and cost-effectively, that moves at least some of the focus from front-end to back-end (from a UX perspective, a blurry division at the best of times, and it highlights the need to forge good working relationships between front/back people outside of big orgs).

WAP was in many ways a word that rhymes with wap. With the rapid adoption of smartphones and tablets and their ever-better displays comes a bigger opportunity and challenge - we know _where_ users are (or at least, their smartphones do).

Location-awareness should let the line between ‘app’ and ‘website’ blur,  and allow some content to bubble up (the contact details, map, timetables, special offers, opening hours, store locator, calendar, today’s blog photo, Twitter feed, share price, etc) - anything that a user might more willingly or readily be seeking to consume on a handheld portable device to meet a time-sensitive or location-related need - be it content to consume on the commute home or instructions on how to find your store. The idea that your content might exist on a user’s screen/in their OS as part of a mashup will also need to grow (iGoogle 2.0, and check out Snippage).

High-value, huge audience mobile tasks can justify the price of dedicated app development, but workaday man-in-the-street sites should be able to afford to provide appropriate and accessible content without having to resort to ‘separates’.

In some ways RWD is web types trying to cope with the awkward teenage years that are inevitable while the fresh-faced app and the self-conscience website grow up and marry as the good-looking, well-adjusted web couple they will become. Even the great BBC is still not there yet - the new desktop approach is ‘swipey-looking’, but the beta mobile site is not that customisable (yet). Once Windows 8 Metro gets here that will impact things again.

Grids and clever CSS/JS can solve the issues around screen sizes, resolutions, ratios, even bandwidth. The whole ‘there’s an app for that’ is out of the reach of the vast majority of small businesses.  What we need is not one website or two, but a website that (metaphorically) does something like this:


Perhaps soon we’ll have easy web levers and switches like ‘mob-priority-1’ or ‘mob-omit’, ‘enable-voice-search’, ‘auto-paginate-swipe’, where these user-friendly choices are being triggered by the OS in the user’s mobile/desktop device talking to the server (and using web storage) and factoring in GPS data. We’re seeing the beginning of this integration with things like geofencing.

So it will I hope one day be one ‘web package’ that’s developed, then dynamically delivered, beautifully tailored, to the user’s situation. In the mean time, whatever Jakob would like to preach, the devil has all the best tunes (and no blue links).

Tim Ball


Interesting, I’ve been doing quite a few parallel desktop/mobile sites of late having despaired of trying find a compromise that works well for image led portfolio sites. It’s an encouragement to look closer at responsive web design, but its going to be an uphill struggle to bridge the gap between the 27” desktop and 3” phone.



Thanks very much for your comments.

Adrian: I’m still digesting some of your very good points - food for thought indeed, and I’ll be checking out the suggestions and links you’ve posted. Thanks. I certainly think that the RESS approach - responsive web design and server side components - is the best way forward. Plain RWD is good for small budget projects, but I’m already facing its limitations when trying to do something a bit more ambitious. There are some fine RESS resources scattered on the web, but I’m pining for a book on the subject. I don’t have the time or knowledge to write one!

Tim: Yes, image heavy sites are a nightmare with RWD. I agree that in the case of photography websites it might be better to stick to the two site approach. Some ingenious responsive image options are available, but nothing wholly reliable yet. I’m trying one or two of them out as time allows.



Of interest re this challenge. And Tim/Justin, if you’ve not seen it yet, check out adaptive-images.com



Thanks Adrian. Viewport units look very interesting. Haven’t checked out adaptive-images.com for a little while, will need to take another look.

Leave a comment

Comments are no longer available for this post.