Is This The Slide That Could Kill Responsive Design?

mobile youtubeResponsive design is supposed to make everyone’s life easier: build once and publish across platforms and devices. Google likes and recommends it because the company only needs to crawl once for responsive sites (Google also supports other approaches). But Google also values fast-loading mobile sites and will begin to punish slower mobile sites.

The slide below appeared in The Search Agency’s report today evaluating the mobile sites of multichannel retailers. It shows how much slower responsive sites load vs. dedicated mobile sites (even PC sites). Update: see second postscript below.

It was suggested to me that responsive design could be done in better ways to address this problem. That may be true. But if the slide below accurately reflects the load-time gap between responsive and dedicated sites there’s a major problem.

Responsive design load times

Not only will mobile site rankings suffer but so will brand perception. Google and others have produced survey data that reflect consumers have little tolerance for slow-loading mobile web sites. There are also well documented brand perception and purchase-intent consequences if sites perform too slowly. (However no one has yet shown that with behavioral data.)

A 2012 study sponsored by Google found the following attitudes and potential business impacts from slow-load times for mobile sites:

  • 72 percent of users said that mobile-friendly sites were important to them
  • 74 percent of respondents said they’re more likely to revisit mobile-friendly sites
  • A majority of users (67 percent) are more likely to buy or convert after a visit to a mobile-friendly site, while the opposite is also true of a non-mobile-friendly site: 61 percent say they’ll “move on”
  • 55 percent said a frustrating experience on a (mobile) website would hurt the perception of the brand

I’m not technically knowledgeable enough to comment on what might be done to speed up responsive design. But it seems to me that it needs to or it may be abandoned when marketers experience the double-penalty of lower rankings and negative consumer response.

Postscript: A number of people have strongly objected and disagreed with this article. I’m not trying to argue from a technical perspective that responsive design is in fact fatally flawed. I don’t have that knowledge, nor a vested interest in the outcome of this discussion.

I was extremely surprised by this slide and wanted to stimulate discussion. However if these results are representative more broadly of how responsive design performs it is in trouble. Let’s be clear, the issue is speed and page-load time.

There is plenty of room for discussion and debate about whether responsive design is being implemented correctly, needs to be modified or replaced by some alternative. I welcome those comments below. Search Engine Land and Marketing Land may do more articles in the future on this topic from a more technical perspective.

Postscript 2: I went back and reviewed the report again. The Search Agency actually recommends responsive design as a best practice. I had been so stunned by this slide that I neglected to see it in the larger context of other slides in the report.

The chart above doesn’t reflect that only 1 percent of the sites examined were using responsive design vs. dedicated mobile sites (or just PC sites). So the sample is definitely too small to be generalized more broadly.

Related Topics: Channel: Mobile Marketing | Top News

Sponsored


About The Author: is a Contributing Editor at Search Engine Land. He writes a personal blog Screenwerk, about SoLoMo issues and connecting the dots between online and offline. He also posts at Internet2Go, which is focused on the mobile Internet. Follow him @gsterling.

Connect with the author via: Email | Twitter | Google+ | LinkedIn



Marketing Day:

Get the top marketing stories daily!  

Share

Other ways to share:
 

Read before commenting! We welcome constructive comments and allow any that meet our common sense criteria. This means being respectful and polite to others. It means providing helpful information that contributes to a story or discussion. It means leaving links only that substantially add further to a discussion. Comments using foul language, being disrespectful to others or otherwise violating what we believe are common sense standards of discussion will be deleted. You can read more about our comments policy here.
  • http://convertwith.me/ flywebguy

    “[responsive design] may be abandoned when marketers experience the double-penalty of lower rankings and negative consumer response.” — There is absolutely not going to be a penalty for being responsive, only a penalty for being slow. Everything we build at my agency is responsively designed and across the past 10 or so sites we’ve built, the average load time is around 3.5s… I think the reason why responsive sites are slower is because the designers are being sloppy. Google won’t ding everyone using RWD because a group of designers are sloppy… they’ll just ding those sites.

  • http://convertwith.me/ flywebguy

    “[responsive design] may be abandoned when marketers experience the double-penalty of lower rankings and negative consumer response.” — There is absolutely not going to be a penalty for being responsive, only a penalty for being slow. Everything we build at my agency is responsively designed and across the past 10 or so sites we’ve built, the average load time is around 3.5s… I think the reason why responsive sites are slower is because the designers are being sloppy. Google won’t ding everyone using RWD because a group of designers are sloppy… they’ll just ding those sites.

  • http://www.mytrialreviews.com/ Sorina Dascalu

    There is also an aspect that nobody takes into consideration when conducting surveys, but I heard many people complain about this: Lack of similarity between desktop sites and mobile sites.

    I heard many complaining that when they access the mobile version of a website they are “lost in space” because they were accustomed to the desktop version and the mobile version is very different.

    This is where responsive design in helping. I wouldn’t ditch responsive design yet, but instead I would work on making it loading faster.

  • Andreas Mitschke

    …which is the #1 reason for a few to ditch the “different” mobile responsive/adaptive site and stick with the “touch screen users are used to pinch-zoom, hence let them pinch-zoom and don’t expose them to a totally different mobile experience – just give them the desktop version, they feel familiar with” understanding.

    - Every page should be optimized to load in ~3s or less, not just a mobile solution.
    - Every page should be below 1MB of ressources, not just the mobile solution.

    Even with increasing bandwidth, there’s no sense burried in not optimizing the desktop version as far as possible.

    Get a real designer who knows how to implement UX architecture into a usability-conform UI.

    Like marketingland’s page for example:

    - crap usability and UX
    ->loaded with 3rd party js widgets

    ->a two-tier navigation, which is totall unrelated and thus counter-intuitive to use – I need to think about it, to understand it.

    ->a package of 3, yeah 3, “related content” parts only “below the content”
    –> a 2×2 matrix content only
    –> a tabbed content window
    –> a freaking big-picture slider

    -> There are 5 Social Media parts cluttering the sidebar, headline trigger, footer

    You don’t need to be an excellent front-end designer and developer like me, to get the point… there might be a lot of “speed” potential, without losing any, “ANY”, SEO impact.

    Just sayin… there’s no need for a specific mobile layout, especially regarding most of the devices feature resolutions, which are en par with business laptops.

  • http://www.mobilemartin.com/ Michael Martin

    Responsive Design is the FLASH design of mobile – easy to setup and seems like a panacea like FLASH was, but misses the mark in intent, loads slow and doesn’t work well across devices.

  • http://www.mobilemartin.com/ Michael Martin

    Replace responsive with FLASH in your statement as I remember when FLASH designers said the same thing many years ago ;)

  • http://convertwith.me/ flywebguy

    http://gtmetrix.com/reports/convertwith.me/oHzxg5VJ
    Your argument is invalid! ;)

    If you’re going to make random, baseless statements – try adding data to seem more credible. Oh, and your website looks bad on mobile and desktop… you might want to fix it.

  • http://convertwith.me/ flywebguy

    I just compared your none responsive website that doesn’t have a mobile version to my responsive website. You can see the results here:
    http://gtmetrix.com/compare/oHzxg5VJ/nbejtX06

    At 5.14s load time, your site is over 2 seconds longer on the load than my responsive website. Trust me… Google likes my load time, and visitors like the experience better.

    Not trying to be hurtful, just trying to bring facts instead of conjecture into the discussion.

  • http://searchmarketingwisdom.com alanbleiweiss

    Thanks for this article Greg.

    Add in the fact that every site I’ve audited that has responsive design has gotten their organic SEO screwed up (all for different reasons), and this is a combination recipe for “WAKE UP DEVS AND DESIGNERS – that “PRETTY” site is no better than FLASH if you don’t pay attention to processing speed or crawlability, and if you don’t properly test across all platforms”…

  • http://searchmarketingwisdom.com alanbleiweiss

    Michael is correct. And my audits over the past year PROVE that a poorly executed responsive design code set KILLS SEO. Your agency may be the exception in this case. Good for you. However to randomly claim there aren’t consequences based on that knowledge is dangerous when others need to learn and know the potential problems.

  • http://convertwith.me/ flywebguy

    I’m not keen on random statements that “prove” stuff without backing it up… care to share your audits to back up your claims?

  • http://battleoflongtan.reddunefilms.com/about/ Martin Walsh

    Are you serious? You have no idea what you’re talking about. Wow, comparing FLASH to Responsive Web Design. Now I’ve heard everything, ROFL.

  • Benjamin Cook

    Replace “responsive” with “web” or “html” or “wordpress” and it’s all the same.

    Bad code can harm your site. Do people run around claiming HTML will harm your rankings?

  • Benjamin Cook

    Alan, there are consquences for any bad code. Claiming “responsive design is bad for SEO” when you really mean “bad responsive design, much like any other bad code, is bad for SEO” is the irresponsible and dangerous statement.

  • http://battleoflongtan.reddunefilms.com/about/ Martin Walsh

    Nice link bait title Marketing Land.

    Poor coding is just that, it has nothing to do with Responsive Web Design and everything to do with Responsive Web Design and ANY web coding in general.

    A lot of website; design and build for retailers and many other brands is done by agencies or internal teams who usually have no experience or skills in website/code optimisation for PC browsers, let alone mobile devices. They also usually have no experience or skills in SEO relevant to today’s best practices.

    I’ve seen this time and time again and it has consistently been supported by reports and survey’s over the past few years.

    To write a very disingenuous headline which blames Responsive Web Design for poor coding is really disappointing on a site like yours.

  • http://convertwith.me/ flywebguy

    You’re confusing correlation with causation. For one, just 4 days ago on Google’s Webmaster Youtube channel, Matt Cutts debunked the worry that responsive design affects SEO (see here: https://www.youtube.com/watch?v=D03wRb4s7MU).

    RWD plays no part in the conversation here… and your flash analogy is bad too. You do know that flash doesn’t even show on mobile devices 99% of the time, right?

  • http://convertwith.me/ flywebguy

    1+ for you, Martin Walsh, on calling a spade “a spade”.

    Matt Cutts just debunked the RWD being a hinderance to SEO just 4 days ago: https://www.youtube.com/watch?v=D03wRb4s7MU

  • http://searchengineland.com/ Danny Sullivan

    Martin, the slide comes from a survey of load times of 100 top retailers. It found that the load time of responsive sites was much higher than separate mobile sites. Those are the stats. There’s no “blame” assigned in reporting those. Maybe the blame is because of bad coding, or better coding that could be done, so that responsive sites load faster. If so, that doesn’t negate the fact that anyone who is thinking they want to go responsive probably also ought to think that they are doing so in a way that ensures the site loads quickly that way, wouldn’t you think?

  • http://battleoflongtan.reddunefilms.com/about/ Martin Walsh

    Danny, then why write a disingenuous link bait headline like ‘Here’s the Slide That Could Kill Responsive Design’?

    The same poor site load times has been evident for the past few years through to today for just the normal desktop based websites for most retailers and brands.

    But, I haven’t seen any hysterical headlines like ‘Here’s the Slide That Could Kill Websites’? on your site.

    It’s about cause and effect. You don’t address the effect without addressing the cause and the cause here is poor coding and optimisation whether that is desktop, mobile or responsive website design. From all the reports, studies and survey’s I continue to see, the majority of desktop versions of websites are still just as bad today from a site speed perspective and in fact it’s gotten worse the past 2 years.

    As I said, this has nothing to do with Responsive Website Design and everything to do with poor website coding and optimisation whether that is for the desktop or mobile devices.

  • http://giacomoballi.com/ Giacomo Balli

    ….just like anything, if implemented incorrectly the benefits will suffer.
    Technically speaking, the problem is that most companies implement RWD by doubling the code/Assets or just adding a checks with javascript… basically half baking their efforts.
    The result is the slide above.

  • http://www.maxminzer.com/ Max Minzer

    I am a huge brand/corporate and have issues with my site speed, website design, information architecture and a ton of other issues. I heard people talk about how awesome responsive web design is and how it solves so many issues and now implemented it on my site…

    Is that the mindset we want to support here?
    This is NOT the issue with responsive web design. And those of us working with brands big and small should take time to educate. Responsive web design is not a magic bullet, just like SEO isn’t. We should know better…

  • http://www.mobilemartin.com/ Michael Martin

    Correct, so we can bring facts, we should actually compare the mobile version of my site vs the desktop instance you chose – it loads over a second before yours using the same tool http://gtmetrix.com/compare/oHzxg5VJ/nbejtX06/vJGH5mwH

    Regardless your site is MUCH better overall and my site is just a barely updated bio site :)

  • Hudson Hornick

    The problem’s not the responsive nature of the design, but the loading time of your site. Sure, responsive design might take longer to load, but done well even large pages load ~3 or 4 sec. and that’s further reduced with asynch ajax, sprites, and consolidating your js and css.
    If your page loading time is ~10 sec., you’ve got bigger issues, partner.

  • http://www.mobilemartin.com/ Michael Martin

    People sure don’t like hearing “their baby is ugly” :)

    Posted about this here at SEL over a year ago http://searchengineland.com/responsive-design-alone-is-not-mobile-seo-124202 and here are some data points I presented at the recent SMX Advanced http://searchengineland.com/4-mobile-search-trends-tackled-at-smx-west-2013-151657

    The point with the FLASH comparison is that just as people were jumping on that bandwagon like RWD, not realizing over time the negative consequences to user intent and search differentiators (mobile in this case).

    GTMetrix is no authority analysis about mobile search which is the context of my argument.

    BTW my site is just a barely updated living resume site that I never bothered with user agent detection to the mobile instance which is at m.mobilemartin.com – doesnt have the higher traffic that your sites do to justify it :)

  • http://www.mobilemartin.com/ Michael Martin

    I can understand why you are defensive, since Alan and I’s points with FLASH are more to do with its negative search aspect than a design argument.

    Also you do know RWD doesn’t work on feature phones, which is important in international markets, fact comes directly from Google https://developers.google.com/webmasters/smartphone-sites/feature-phones

    Your citation of Cutts saying RWD isn’t a disadvantage, is true since its really “no harm no foul” and why the mobile crawler doesn’t bother with it since its just the same HTML, so its not an advantage either.

    Matt did another video saying Google’s mobile crawlers are interested in dynamic serving’s separate device HTML pages by properly declaring the Vary HTTP header http://www.youtube.com/watch?v=va6qtaiZRHg – this could lead to an actual advantage within mobile Search.

  • http://searchengineland.com/ Danny Sullivan

    It’s a headline encouraged to get people to read and discuss. That seems to have succeeded. A survey finds that responsive sites are slower than even desktop sites. Assuming the survey is correct, that’s an issue that should be solved. That’s not to say, and the headline still doesn’t say, that the “blame” is on responsive design. It’s just telling you that with responsive design, there might be slowness issues — so you’d better correct those. And the reason why you don’t see this about web sites in general is because (1) that’s been said already plenty in the past and (2) this slide showed responsive sites coming in worse.

  • http://www.mobilemartin.com/ Michael Martin

    Great common sense comment :) Dealing with Fortune 500 companies we agree RWD fits with most pages on scale when the intent doesn’t vary by device, but when it does, dynamic serving makes more sense toward the return in investment

  • http://searchengineland.com/ Danny Sullivan

    No one said a responsive site was a hindrance. Not even in this article. It said that a slow site might be a hindrance, because Google — including Matt himself — says that slow sites might suffer a penalty. Matt’s video on responsive design isn’t addressing slowness. He’s talking about mainly URL issues and how you might get PageRank credit somehow being lost, and how responsive might actually help there. But at 1:37 he specifically says that you should worry about whether “it’s fast.” So if you go responsive, but you’ve done that in a way that makes your site slow, you might have an issue.

  • http://searchengineland.com/ Danny Sullivan

    See above. As I commented about that video, he specifically says in terms of sites that you’d better make sure they are fast.

  • http://searchengineland.com/ Danny Sullivan

    The story never says there’s a penalty for being responsive. it says that this survey found responsive sites were slow. So the story suggests that if you are building a responsive site, you’d better make sure you keep it fast — because if you build a slow site (responsive or otherwise), you might have an issue.

    I’m glad you’re building responsive sites that are apparently fast. But clearly, others are not.

  • neotrope

    We have a 512MB load responsive design website which loads in under 2 sec, with javascript, photos, logos, bootstrap, font awesome, and google fonts. Of course we’re using shtml, Amazon S3/CDN, and compressed the heck out of everything. Page Speed 92 / Yslow 96 (was 94/98 but I have an element which requires caching with local versioning and get penalized for a ? on the cache element …. so, proper design with responsive will work. With the “average” page speed being “76″ or something, anybody not hitting 80 for mobile and 90 for desktop will simply … FAIL. :-) (just commenting not trying to start a long winded thread … I’ve spent 3 months tweaking the crap out of every little thing to get t he responsive template rockin’ … and it has hurt big time …!) But now, I feel comfy we don’t need to have both an m.mysite and mysite.

  • http://igvita.com/ Ilya Grigorik

    I believe its worth pointing out that out of the sampled 100 sites in this report, there is only a single (1!) site that was using RWD. As such, I would consider drawing any conclusions on performance of RWD as a whole, based on a single site, as unwise.

    Really, what this headline is saying: we found a slow RWD site, hence all RWD is and henceforth will be slow!

    That said, the site in question could definitely use some optimization work:
    https://developers.google.com/speed/pagespeed/insights/?url=http%3A%2F%2Fwww.carters.com%2F

    - Not compressing images (900KB+ of wasted bytes)
    - A ton of scripts / css in the critical path which can be eliminated

  • covedigital

    Love it, any excuse for a ‘responsive design debate’ fortunately it’s the way forward, but bravo for finding an angle to kick off the debate again, like many have stated (above or below, depending on where the comment sits) it is possible to reduce the load time, by many factors, too many to include in a small comment, add to this once there is a consensus on how best to load retina images further improvements will be made, but as stated well done for started the topic!

  • http://www.nathanielbailey.co.uk/ Nathaniel Bailey

    Sorry but what sort of sites where the load times done on? And what were the differences in servers etc these different sites where on?

    We are currently changing http://www.splashandrelax.co.uk to a responsive design (sorry can’t give the url for it as its still under construction) and the home page which is finished loads just the same as the current standard one!

    I’m not having a dig at you, but these guys that have given these numbers are giving the wrong impression about responsive sites and its not right in my opinion.

  • http://aquariumdrinker.com/ davidconnell

    I don’t know… The second postscript on this story also says “he sample is definitely too small to be generalized more broadly.” I feel like the title, what the article actually says, and then what the source material actually says just don’t match each other and this is just bad journalism. It’s probably a post that shouldn’t have made it onto the site and probably should have been corrected at the top of the article after the author actually read the whole report beyond the “shocking” slide/chart.

  • http://convertwith.me/ flywebguy

    Interesting – when visiting your site on multiple mobile devices, it never sends me to the mobile site. You might check your .htaccess to make sure the rules are correct and cover all devices.

    However, as you’ll notice in the test you linked to, your mobile site is less than 10kb and only loads 3 resources… that’s not a site, its a sentence. Not exactly a comparison now is it? :)

  • http://searchengineland.com/ Danny Sullivan

    There’s a link to the report in the post.

  • http://searchengineland.com/ Danny Sullivan

    Yes, Greg’s added a note there.

  • http://www.mobilemartin.com/ Michael Martin

    Being a bio site than a business site, a sentence covers what little I accomplished ;)

  • http://convertwith.me/ flywebguy

    In the video he didn’t say that it’ll lead to an actual advantage, but that using the Vary HTTP header will allow the various server caching resources to MAYBE infer a benefit. Too much work for very little benefit. Just do server side caching of the resources on your responsive design.

    Regarding your comment about Feature Phones – the average site we optimize has between 20% and 60% of their traffic coming from mobile. Of that, less than 1% is from feature phones. So why would I build multiple m.mysite.com for the various feature phones when they represent less than 1/2% of the overall traffic?! #pennywisepoundfoolish

  • http://www.mobilemartin.com/ Michael Martin

    Skitzzo,

    I agree with your point, but the context of my stating FLASH in the above is how “back in the day” it was the “bees knees” of development and what could go wrong…with obvious SEO implications, slow loading, and device/browser rendering. Same will occur over time with RWD (no MOBILE SEO advantage, slow loading, and doesnt work on feature phones).

    I can detail more when we get together for our SEO panel at Affiliate Summit in a couple months…also don’t be mad about the whole World Series thing and my Red Sox ;)

  • http://www.mobilemartin.com/ Michael Martin

    Correct in the US you really wouldn’t hence why I stated internationally, where say India, still the vast majority of mobile searching is done with feature phones.

    You are also correct on assessing the ROI benefit for the effort of implementing dynamic serving or having a separate instance for an m.

    I stated the justification is dependent on if there is a difference in intent for certain critical pages, else RWD suffices…from a search and user perspective.

  • http://convertwith.me/ flywebguy

    Same for any site – slow kills. But the context of this article is that RWD kills… which is wrong. RWD doesn’t kill anymore than HTML & CSS kills. The author of this article stated that there is a double penalty for slow loading and RWD. Which is simply not the case and why myself and many others have voiced our concern over the lack of journalistic integrity and correct facts reporting.

  • http://convertwith.me/ flywebguy

    Even their desktop site ranked 50/100. That’s horrible. Thanks for pointing this out.

  • http://convertwith.me/ flywebguy

    OK, so case in point – your Covario company site utilizes responsive design but takes 5.56s to load and the page size comes in at 1.22mb with 66 resources. But the problem isn’t with responsive design… if you look at the waterfall graph from a test, you’ll notice that the two worst issues are 1) custom.css and 2) that 1 non optimized image that is ~600kb all on its own.

    This underscores that RWD is no where near the cause of the issue… its bad site optimization in general. Those issues would be the same if Covario’s site wasn’t responsive.

  • http://www.mobilemartin.com/ Michael Martin

    Don’t get me started on our Covario site ;)

  • Harold

    Thankfully no real dev would listen to what you’re suggesting with the title, but it’s a complete disservice to any potential client who might find this and not know any better.

    What you’re recommending is going back to a practice where the client’s mobile site may not load at all (I mean, we’re examining devs who don’t optimize for mobile, so we can also assume their UA sniffing is not up to snuff, right?).

    If anything, you could be educating people on what they should demand from devs in regard to site performance, but that probably wouldn’t be good link bait, right?

Get Our News, Everywhere!

Daily Email:

Follow Marketing Land on Twitter @marketingland Like Marketing Land on Facebook Follow Marketing Land on Google+ Subscribe to Our Feed! Join our LinkedIn Group Check out our Tumblr! See us on Pinterest

 
 

Click to watch SMX conference video

Join us at one of our SMX or MarTech events:

United States

Europe

Australia & China

Learn more about: SMX | MarTech


Free Daily Marketing News!

Marketing Day is a once-per-day newsletter update - sign up below and get the news delivered to you!