Repurposing vs Optimized Design: It’s Not a Battle

Jakob Nielsen of useit.com is no stranger to controversy and his recent Repurposing vs Optimized Design article makes a bold statement:

It’s cheap but degrading to reuse content and design across diverging media forms like print vs online or desktop vs mobile. Superior UX requires tight platform integration.

Jakob states there are two opposing schools:

  • Repurposing (responsive design): make as few designs as possible and reuse the same material across platforms. This has cost advantages, but results in a substandard user experience.
  • Platform optimization (device-specific design): create different user interfaces for each platform and integrate the user experience layers as tightly as possible. The result: higher costs but better usability and bigger profits.

The article resulted in a Tweetstorm:

Jakob Nielsen analyzes Responsive Design and comes up with dumb conclusions. Again.

For 17 years, Jakob Nielsen has been citing his own work as proof of everything. Somebody make it stop.

I’ve thought Jakob Nielsen was a self-important snob for years. Glad to see the Cult of Nielsen crumbling at last.

Do people still pay any attention to Jakob Nielsen or is he hanging around just to annoy web developers?

Jakob Nielsen is 100% right on mobile. Says Jakob Nielsen.

It’s amazing to watch Jakob Nielsen make himself more and more irrelevant with almost every piece he writes.

The Weaknesses of Responsive Design

Jakob claims most responsive websites are too primitive. They rearrange the same content and change dimensions of some elements, but rarely:

  • Adapt the content, i.e. provide shorter and simpler articles for mobile.
  • Use alternative, cropped or zoomed images.
  • Cater for touch input.
  • Reduce or modify the feature set to suit smaller devices.

Websites should consider the user context. Requirements change depending on where a user is, what they’re doing and what device they’re using. Consider the website for a bus company; a desktop user may want to browse timetables whereas a mobile user may need to find the nearest station.

Jakob admits this could be achieved with responsive techniques but, once you account for all these differences, you effectively have two separate designs. But he makes a valid point: if potential revenue is greater than cost, a site designed specifically for mobile devices makes good business sense.

Calculating Potential Revenue

This is where Jakob’s argument breaks down. For most websites, the cost of developing a separate mobile site makes no commercial sense whatsoever. This is especially true when you consider Jakob’s suggestion of creating and maintaining two or more sets of content.

The majority of websites receive few mobile users — usually less than 10% of visitors. It may be increasing, but does it make commercial sense to throw money at a problem which does not yet exist and is subject to radical technical changes? I’m all for being pro-active, but this leads to another point: you cannot make assumptions about user a user’s habits, desires or requirements.

Jakob paints a black and white picture: mobile users want less content and simpler features. However, it’s impossible to determine context when all you actually know is the screen resolution. The situation is made more difficult with newer mobiles which have screen sizes approaching those of tablets and the relatively diminutive iPad 3 exceeds the resolution of a top-end desktop PC. You may be able to make context judgments based on the current time or GPS location but, ultimately, you don’t know whether the user is rushing for a bus or lying in bed.

The Strength of Responsive Design

Responsive design provides a solution. A website primarily designed for desktop use can be restructured to provide a mobile-friendly experience at minimal expense. It may not be perfect, but it opens access and your site isn’t restricted to certain devices or user groups.

It’s a starting point — not the end. It’s now possible to gather statistics about devices and tie them to typical user journeys. If you can establish that X% of access is from a smartphone and Y% of those users proceed to feature Z, you may be able to justify further development. In extreme cases, a standalone mobile site will aid usability and increase profitability.

It’s Not a Battle

Jakob makes two broad assumptions but the reality is not as clear-cut:

  1. You have a mobile context or a desktop context.

    There are thousands of differing devices, screen sizes and situations when they are used. People use their mobile while watching TV — should they always receive a summarized article? Mobile exceeds desktop usage in many African countries because it’s cost-effective and reliable — should they be offered restricted functionality?

  2. Repurposing and optimizing designs are opposing technologies.

    Why? There are an infinite number of ways to develop the same solution. An optimized design can still use responsive techniques and vice versa. Ultimately, either can be used in differing quantities to achieve the same goal.

Jakob has been accused of bashing responsive design. The tone of his article gives that impression and it detracts from his main point: a dedicated mobile design offers a better mobile experience. If he’s guilty of anything, it’s stating the obvious.

Free book: Jump Start HTML5 Basics

Grab a free copy of one our latest ebooks! Packed with hints and tips on HTML5's most powerful new features.

  • http://www.useit.com Jakob Nielsen

    Thank you for your comments on my article.

    I am not quite sure what you mean when saying “This is where Jakob’s argument breaks down. For most websites, the cost of developing a separate mobile site makes no commercial sense whatsoever. “

    I believe that was exactly what I was saying in a rather long article on the cost-benefit of repurposing vs. platform-optimized design. Maybe I didn’t say “most sites,” but I definitely said “In many cases, the cost–benefit analysis favors a unified design.” (The full 293-page report from our mobile usability studies has 4 pages specifically about the question “do you even need to consider mobile in the first place”. I admit that this is only 1% of the report, but then again we do write for the target audience for each publication, and most people who get this report do so because they want advice on how people use mobile user interfaces – and thus how to improve them.)

    Anyway, if our main disagreement is between “most” and “many”, it’s not so bad :-)

    One last point: it’s not true that I have been citing my own user research for 17 years. I am actually on my 30th year now of relying on empirical testing with real users as the background for my design recommendations. I wish other people would do the same.

    • http://www.redubl.com Dustin James

      “One last point: it’s not true that I have been citing my own user research for 17 years. I am actually on my 30th year now of relying on empirical testing with real users as the background for my design recommendations. I wish other people would do the same.”

      This. 100 times this.

      Too much of this industry is based on anecdotal arguments as well as people taking the results from one case to be cause and effect. Behavior does not work that way – we need large samples, validity, reliability and replication.

      Also, huge respect for responding to a blog post on your recent article. Fantastic.

      Dustin

    • http://www.optimalworks.net/ Craig Buckler

      Thanks for your comments, Jakob. Your article makes sense and I agree with what you’re saying but, in some instances, it comes across as anti-responsive design. I doubt that’s your intention, but terms such as “degrading” and “substandard user experience” were certain to provoke a reaction. (Admittedly,Twitter amplifies the volume of debate.)

      Responsive design is not particularly new; fluid layouts have been with us since the start and are one of the medium’s best strengths. (They only fell out of favor because designers found them difficult to comprehend.) RD is not always used well, either; some designers simply opt for a series of fixed-width layouts.

      However, when used appropriately, RD techniques offer a baseline mobile experience which allow companies to assess mobile usage and make commercial judgments based on facts rather than feelings. Your article states businesses should perform cost-benefit analysis but doesn’t consider how that can be achieved. In some cases, responsive design is a logical first step for build-evaluate cycles.

      Ultimately, company research may result in a separate mobile site. But that doesn’t preclude responsive design techniques being applied for different mobile screen sizes and orientations.

      • Vince

        RWD, as a concept, is two years old tomorrow. There are missing components (images! woe to images!), there are misuses (…sometimes too numerous), and there are misunderstandings. Give it room to breathe; give the fanatics room to be, well, overzealous, and give opponents room to get over the inevitable shift.

        And not to throw shade, but sometimes when we devs are resistant to embrace new tech and new methods, I’m reminded of the whole “tables vs divs” shift.

    • itmitică

      I, as a customer, not as a developer, would consider your report the moment you do, and not a second sooner. How about proving you’re money’s worth and actually implement some of your findings in your business website? It, in all particularity, doesn’t cover any of the mobile increased usability, mobile different content or mobile different design.

      I will, once again, outline the difference between responsive web design, which is style and mobile web sites, which are alternative content. Totally different concepts. The confusion should stop and it should be very clear to anyone that responsive web design pays to be used always, on “normal” web sites as well as on their alternative mobile web sites.

  • http://www.redubl.com Dustin James

    Good article Craig. I like your analysis of Jakob’s analysis.

    Two points for consideration:

    1.) Separating the marketing from the input cost of design. True, mobile users account for approximately 10% of all web traffic. However, I have seen business cases where mobile conversion rates doubled or even tripled the conversion rates of desktop users (over considerable sample sizes). Most assumptions would be that people won’t buy as readily on a mobile device as a desktop, so in short, you’re right – we shouldn’t assume anything about users behavior. In these cases, responsive design might not be a good solution, so the question becomes – is it one of those cases? Hint – you must test to answer, guessing sucks in this business.

    2.) Sometimes, too much is assumed by the design community. Personally, I see the merits of both responsive and device specific design. Ideally, both are case dependent. However, you rarely know until you launch, gather data and analyze. This is the trouble with trying to provide ideal solutions for a fixed cost. So we’re left to guess and hope we’re right with our choices. Maybe we need to sell web solutions as a process rather than a one time product delivery?

    In any case, I personally like Jakob. I think that, for the most part, he backs up his arguments and analysis on facts and data. Most arguments against tend to be anecdotal. John Meynard Keynes said “my mind changes when the facts change.” 140 character rebuttals don’t really sway me :)

  • http://wpconsult.net Paul

    there’s been a lot of negative attitude recently towards anyone who dares say anything negative about responsive design.

    • http://www.redubl.com Dustin James

      No kidding, I feel like I’m weighing in on a religious debate

  • itmitică

    Responsive web design === CSS.

    Mobile web sites === Content.

    I’m pretty sure any mobile web site (content) can actually benefit from a responsive web design layout (CSS). As any content benefits from a dynamic layout.

    Think about: switching portrait to landscape and vice-versa.

    Think about the variety of mobile display sizes: 240px up to 1024px… and beyond.

  • http://www.linkedin.com/profile/view?id=14898459 Grant Parks

    I agree with your thoughts about assumption #2. Architecture emerges and evolves. Where someone starts out on this continuum of strategies should be dictated by the current state of their application as much as anything else. It’s not a question with one answer for everyone. To apply a “right answer” to every situation will incur extra costs and risks, and amounts to taking the easy way out to avoid analysis.

  • http://sultanshakir.com Sultan Shakir

    I’m shocked that Jakob took the time to respond to this post on Sitepoint. But then again, if I was in the business of selling controversial research, then comment spamming a link to my $298 report might not be a bad idea. The worst that could happen is that my link could be pulled down and I’d look desperate.

    Next time maybe he should ask for an interview or send the a letter/email of appreciation for the press.

  • http://www.webvanta.com Michael Slater

    I find it strange, but not entirely surprising, what a religious debate this has become.

    Too many people think mobile-optimized = responsive, and they think they need a responsive site when the really need a mobile-optimized site. Responsive design is one way to get there, but not the only way.

    It is true that you need to implement fluid designs for mobile-optimized pages, because of the different sizes of phones and the portrait/landscape issue. However, this does not mean that these pages need to be the same ones that are used for the desktop site.

    I use responsive design for sites where the content layout is pretty simple and you want the entire site to be available on mobile devices.

    I find that the responsive approach falls short in some cases because of the following issues:

    1. It encourages designers to think about the mobile design as a derivative of the desktop design. Often, for a complex site, you want something completely different for the mobile design.

    2. It doesn’t readily support use of mobile frameworks such as jQuery Mobile, which can be big productivity enhancers.

    3. It often leads to complex HTML and CSS that can be difficult to test, debug, and maintain.

    Here’s the article I wrote on this topic a couple weeks ago:

    http://www.webvanta.com/post/812351-responsive-design-is-not-the-only

    • http://www.optimalworks.net/ Craig Buckler

      I agree that RD does not necessary equate to mobile-optimized but, in response to your points:

      1. The mobile-first method of coding (which I recommend) considers desktop design as a derivative of mobile design.

      2. You require a separate mobile site to use jQuery Mobile. That’s the point of it — you could never squeeze it into a responsive template.

      3. Your HTML shouldn’t be affected to any great degree. CSS will be a little more complex but, assuming it’s done well, it only requires a few additional styles which can be defined in separate CSS files. Besides, how is that any more difficult than testing, debugging, and maintaining two distinct sites with separate content and templates?

  • http://jrwdev.com Jonathan Wadsworth

    ” I’m all for being pro-active, but this leads to another point: you cannot make assumptions about user a user’s habits, desires or requirements.”

    This seems true at face-value. However, those who do and happen to end up right become trendsetters and billionaires (e.g. facebook).

    • http://www.optimalworks.net/ Craig Buckler

      Facebook is a case in point. They throw hundreds of features and applications at user. Some work, some don’t. They’re testing all the time — they’re not making assumptions about how, when or where users will interact with the system.

  • http://gregpettit.ca Greg

    Certainly it’s only being presented as an option to consider, but I would never consider reducing content for the mobile user. Being on mobile doesn’t preclude the desire to read an entire article, tutorial, story, or whatever. Image size and other visual niceties, sure (though some people have higher speeds and more data on their mobile plans!). But content, no.

    • http://www.optimalworks.net/ Craig Buckler

      I think it works in certain situations, e.g. rather than showing a link to an article using it’s title and summary, you omit the summary or make it shorter. But you’re absolutely right — ultimately, a user will want to read the full article and it shouldn’t be hidden from them.

  • Mark

    Good arguments all around. I think one shouldn’t get too emotionally attached to things or ideas like ‘responsive design’. I also think Mr Nielsen does not have ‘controversy’ in mind when he makes a ‘bold’ statement. Just use the best available tools/techniques/ideas wisely. Ten, fifteen years from now, who knows what will be next big UX idea?

    Having said that, I do agree that ‘superior UX requires tight platform integration’–just look at Apple’s products for one–but perhaps the breadth of technology and the economic realities today (there would be more, I gather) are just blurring the criteria for or against repurposing vis-a-vis platform optimization. Nothing to get upset about really, but something to think about. Now if I can only afford that NNG report. :-)

    • http://www.optimalworks.net/ Craig Buckler

      Microsoft created tablets, PDAs and mobile phone interfaces many years before Apple, but they were never successful. The main reason: they tried to squeeze the desktop OS onto mobile devices. Windows is the world’s most dominant desktop platform, but that doesn’t mean its metaphors work well in all contexts.

      Apple started from the ground-up. They addressed the differing needs of mobile, tablet and desktop users and have become the world’s largest IT company as a result.

      However, Jakob’s comparison of Apple and Microsoft is an extreme example. Most of us have sites which are predominantly accessed by desktop OR mobile users — it’s rarely 50/50. Responsive Design allows us to support both groups at minimal cost from day one. It may not be perfect, but can help us assess what users want and make good commercial decisions about the future direction of the product.

      • camus

        MS failed because unlink Apple they did not deliver content. no itunes , no appstore , and it was a time when internet on mobile was too expensive to use. And there were performance issues too. In the end windows mobile 6 was a good os , just that it was hard to find/get associated content or new apps for that os.

        About responsive design , it not the answer , but it is better than nothing when you dont have the resources to developp an entire mobile version for your website. anyway good article , thanks

  • http://gauravmishra.com Gaurav M

    Responsive Design is a improvisation

    • itmitică

      Actually, what sounds more like an improvisation, is building different variants of content for the same website.

  • http://www.idude.net Chaz Scholton

    My own personal opinion about this topic. I would hope that developers would not toss aside their own experiences and reasoning skills. Personally, from all my experiences I’ve come to embrace the concept of “Just good enough” compared to seeking code Nirvana. The persuit of Nirvana (perfection) can drive up time and cost.

    Is Jakob Nielsen right? Yes, but only in some cases. Is Jakob Nielsen wrong? Yes, but only in some cases. Truth be known, what is the case you as a developer are faced with? What is the budget for the project? What are the resources?

    I’m always amazed by websites that are Making money, yet the code is not in a state of Nirvana. It does not have to be the latest high end cutting edge site. It just has to work and function in a manner that allows users to achieve their goals. User Goal driven design.

    If you have a crappy user interface design, crappy content, if your clients don’t have good business practices, if you clients even suck at marketing. All the coding you do in the world, won’t change it. In fact, you can end up adding onto a Tower of Bable to no avail.

    I think time and time again, the concept of KISS (Keep it Simple Stupid) works the best.

    The Content for many websites is stored in database, files or rather isolated from the website application code itself. It’s easy to tap into and display in hundreds of different ways.

    I think the issue here is centered more around the various frameworks/platforms which have been or will be implemented. There are frameworks for everything now days.

    An ecommerce website built using Miva Merchant (back a few years ago) which is in operating and making money for the client verses an ecommmerce website built using Magento. There’s a bit of a difference here when it comes to expanding or making modification to the code.

    This is where reality can be a bitch slap in the face. Database structures of both ecommerce software solutions radically different even. The SQL involved in fetching data can become rather complex unless you understand these systems inside and out. So simply creating a second website to display this content is not very practical nor cost effective.

    However, for some websites the database design and application logic is much simpler. One can create a second site designed for mobile very easy.

    Personally, I’m a big fan of reusing code especially SQL and Application Logic. If I need to fetch records, update a record, delete a record…(the whole CRUD concept), I simply love to tap into same code for it. Why reinvent the wheel again?

    The point I’m trying to make, is it all depends upon how complex the website is. Complex in terms of User Interface, features…and complex underneath the hood at the code level.

    When I read debates like this, I end up cringing at my keyboard. Do I think Jakob Nielsen is right or wrong? My response is that it all depends (show me what the specific project is and I’ll get back to you).

    As developers, programmers, geeks that we are… there’s a lot to be said for dealing with things on a case by case basis. Also, to what ends are be trying to perfect Nirvana? The shelf life of Nirvana is not very long these days. Everything we are coding today will become outdated just as fast as the computers and tools we use to code with (seriously this is reality).

    I do this, I’ve had a hard enough time to convince clients to make the additional investment into content creation. I can imagine how this will be an even more difficult sell, advising them to create special more condensed copy just for mobile. However, some clients have volumes of content created and won’t have an issue in producing condensed copy for mobile.

    My own experience is to expect users, clients and others to be lazy. Everybody wants the world to be a simpler place instead of a more complex one. However, be certain to never exclude anybody which is not so lazy. i.e. Advanced users, clients and others.

  • dave

    There’s no arguement that multiple websites are better than one responsive one. Is there?
    A bespoke website for a specific screen sizes with bespoke features and content is bound to be better than a generic one that just adapts to every device.

    Better for the user, that is. The trouble is, it costs a lot to develop and maintain, and it’s not like we are developing for 1 phone, 1 tablet and 1 desktop. There are so many device sizes out there than sites have to be responsive, or adaptive in some way.

    Jakob Nielsen’s article was geat. If only because it raised this debate. Developers need to chill out a bit when someone doesn’t jump on the latest fad (not that responsive is a fad). There is some middle ground. I think Jakob Nielsen knows this.

    Take everything you read with a pinch of salt. Read between the lines. Digest the information and try and look at all sides of the story. Responsive is great in some cases but not in others, thats at least what I took from Jakob Nielsen’s article.

    • itmitică

      “There’s no arguement that multiple websites are better than one responsive one. Is there?
      A bespoke website for a specific screen sizes with bespoke features and content is bound to be better than a generic one that just adapts to every device.”

      Unless you know that responsive web design is what you need on mobile web sites too. Think about the variety of mobile display sizes: 240px up to 1024px… and beyond.

      • http://www.optimalworks.net/ Craig Buckler

        The issue is becoming more complicated when you consider the resolution of newer mobiles. The Samsung S3 offers 1,280×720 – more than some laptop screens. A typical mobile site may look poor whereas a desktop site is unreadable. There is no one solution for all websites.

  • http://webdeavour.co.uk Lee Kowalkowski

    “The majority of websites receive few mobile users — usually less than 10% of visitors. It may be increasing, but does it make commercial sense to throw money at a problem which does not yet exist and is subject to radical technical changes?”

    I don’t know about you, but if I go to a site on my mobile, and it’s not as easy as using a smartphone app, I rarely return to it again. The fact that less than 10% of visitors to a site are mobile might be a measure of quality itself. I don’t see how you can wait for an audience to grow before enhancing their experience, i.e. if the experience is unsatisfactory, the audience will not grow.

    I’m surprised at the reaction to Jakob’s findings, especially considering the number of smartphone apps there are that are essentially bespoke interfaces to existing websites. There doesn’t seem to be the same uproar about their creation being a maintenance overhead or otherwise. Many companies seem to urge mobile users to use their app as opposed to their website.

    • http://www.optimalworks.net/ Craig Buckler

      If your site doesn’t work on mobiles it stands to reason you’ll have fewer mobile users. But, RWD allows you to make a reasonably good mobile experience for little outlay. That may be all you need but, if you subsequently discover that the majority of users are on a smartphone, you can make a commercial judgment about creating a separate mobile site or native app.

      Mobile for some sites/applications makes sense, e.g. Twitter. But it’s far less necessary for others, e.g. PC software providers. Everyone’s situation is different but building a mobile app simply because it’s possible isn’t necessarily a good business decision.

      • dave

        So maybe responsive isn’t the way to go anymore.
        Perhaps we should just build fluid sites that scale from large to small (without media queries swapping css at regular intervals).
        These adaptive sites require more thought (in my opinion) because of things like line length etc but it’s certainly something I’m trying to work with.
        It’ll also account for people who don’t maximise their browser window on medium to large screens.

  • http://simon.fi/ Simon Rönnqvist

    The largest portion of sites that I develop are brochure-type presentations of businesses, and for those there’s no reason what so ever to present separate content for mobile, desktop, tablet or whatever.

    I also don’t think that it’s any easier to read long texts on a desktop/laptop computer than on a decently sized smartphone, actually I prefer reading ebooks on my smartphone instead of reading off the computer screen. Tablets are of course the preferable to for longer texts, but that’s a different story.

    Separate sites make sense when the you want to offer different content depending on the device. For example an event might want to bring up the minute-by-minute schedule in their smartphone version of the site, while the normal site might want to present you with more general information up front… assuming people use it when deciding whether to attend the event at all. In this case they might want to change the mobile site to the minute-by-minute schedule the day the event starts, not before that. Again were making assumptions about the user, even though they might be quite valid ones.