13 Comments

Summary:

Publishing analytics startup Parse.ly has released a report highlighting the top screen sizes on which its customers’ readers are consuming news content. Desktops and laptops still dominate overall, but Apple is king of the mobile devices.

Source: Parse.ly

Analytics startup Parse.ly has released its third bi-monthly “Authority Report” on Wednesday, and the big finding is that the majority of traffic for some top online publishers still comes via laptop and desktop computers.

The company analyzed the screen sizes of devices that its customers’ readers use to access their content, and found that about 60 percent of all traffic came from devices with screen ratios of 16:9 or 16:10, indicating it came mostly from laptops or desktops. Overall, traffic from mobile devices accounts for between 10 and 20 percent of traffic, peaking at nearly 20 percent in the evenings and with most of it — at least 13.8 percent — coming from Apple iPads or iPhones.

Source: Parse.ly

The 2,000 screen sizes driving 99.5 percent of traffic. Source: Parse.ly

This is the third time Parse.ly has released a report on traffic trends. A highlight of its first report was Feedly’s dominance as the reader app of choice in lieu of Google Reader, while its second highlighted the high percentage of traffic coming from so-called dark search. There’s an interactive version of the screen-size chart (and a couple others) available online.

  1. A lot of internet business I will not do when on my Android phone since after 5 days on a new “unlimited” data plan I was informed by text that I had reached my “free” plateau / maximum. All I had done was read one story on a very common news app. I removed it so my other data would not be hampered.

    Share
  2. I don’t know that landscape and 16:9 necessarily means desktop… but whatever.

    Share
  3. Flawed methodology? How many people have smartphones with resolutions that are common to desktops and laptops as well? And how many people read the web with their phone sideways (landscape)? And how many people have at least one desktop monitor set to portrait orientation? It seems like they are trying to infer too much from the information they’ve gathered.

    Share
    1. Andrew Montalenti Tuesday, December 10, 2013

      Hey Shenan,

      If you look at our more detailed interactive display (http://www.parsely.com/authority/2013/12/), you’ll also see a time-of-day mobile-vs-desktop breakdown. This display is based on browser user-agent, not on client device resolution. What you’ll see there is that mobile usage peaks at around 20% and is as low as 10% during the day. One thing is for certain — desktops dominate news browsing habits during working hours (we imagine this is due to people sitting at their desks, reading news on their workstations). Mobile usage spikes up a little bit during nighttime hours and weekends (which also makes some intuitive sense — people are out and about, or lounging at home, phone/iPad in-hand).

      You’re right that the screen resolution data does not separate out desktop/laptop browsers from high-res mobile devices that happen to match typical desktop/laptop resolutions. But, as I mentioned above, mobile is still definitely in the 10-20% range, not at all dominant, and especially during working hours. I imagine this will change over time, but we’re not in a mobile-only world yet, not by a long shot. And the variety of screen resolutions shows how complex the ecosystem is, and why responsive design has some strong basis in reality.

      Personally, I think publishers should take feel free to take advantage of some of the extra real estate to make some immersive journalism for their readers. And given the way phone resolutions are going, it’s clear it will be an “HD world” soon!

      Explore the data yourself: http://www.parsely.com/authority/2013/12/

      –Andrew Montalenti, CTO (Parse.ly)

      Share
  4. Device and OS are in every publisher’s Google Analytics. Why is this a story?

    Share
    1. Mainly because this is an aggregate calculation across hundreds of publishers. Secondarily because screen size is more granular than OS and browser.

      Share
  5. Nicholas Paredes Tuesday, December 10, 2013

    You relied on chance and won!

    The rest of today’s articles I read on my iPad.

    Share
  6. Hmm, well, I’m apparently in the minority here, but I read all of my articles on my Nook Tablet. Works just fine for me :-)

    Share
  7. Just to be clear: how is traffic being defined in this report? Is it bits, usage hours, pageviews? Might the results be different based on different traffic metrics?

    I’d also be interested to know if there’s a geographic dimension to this data. I do a lot of research on digital trends in emerging Asia, which have higher levels of mobile centricity. I wonder how people in places like the Philippines or Indonesia are consuming content…

    Share
    1. Andrew Montalenti Wednesday, December 11, 2013

      Will –

      There is definitely geographic bias in our data, we should have been clearer about that. We work primarily with US-based publishers, but also a few in Canada, Australia, and Europe. We have very few Asian publishers at the moment (hopefully, that’ll change over time!) and so we should probably only treat this as a truly representative sample of the current state of news browsing in the US.

      –Andrew Montalenti, CTO (Parse.ly)

      Share
  8. FTR, I’m reading this on an HTC One phone … but might have also read it on my Nexus 7 tablet (via Reuters Counterparties email). Guess I’m a minority.

    Share
  9. The conclusions here seem reasonable but any time a company publishes ‘research’ that comes from an analysis of their own customers there can be a problem with bias. What do these customers have in common that makes the parse.ly service attractive? Similar target markets, similar business models, similar content? They are NOT a random sample of publishers. The only thing that can be said with any certainty is, “Chances are you’re not reading this on a mobile device if this site is using parse.ly” This is in no way an “Authority Report”, it’s a PR campaign.

    Share
    1. Andrew Montalenti Monday, December 16, 2013

      Kelly,

      There not only “can be a problem with bias”, but there *is* some bias in this data — as there is in any aggregate report!

      Parse.ly customers have in common that most are US-based, most are “premium” publishers (e.g. household name type news brands), and most have audiences of 1M unique visitors per month or more. This is why we feel confident saying that this is a good sample of US news reading habits, not generalizable to all websites, but certainly interesting data for the purpose of running news / reference / information websites.

      We also tried not to draw too many conclusions ourselves, as you can see if you explore the data interactively here: http://parse.ly/authority/2013/12/

      –Andrew Montalenti, CTO (Parse.ly)

      Share

Comments have been disabled for this post