Media Metrix Multi Platform

comScore is announcing something today that I am quite excited about. It is called Media Metrix Multi Platform and comScore describes it this way in their announcement:

Media Metrix Multi Platform offers unduplicated accounting of audience size and demographics that reflects today’s multi-platform digital media environment, which includes websites, apps and video content accessed from multiple devices.

This "multi platform" environment is the reality of most online properties today. We access them on the desktop/web, the mobile web, and on iOS and Android apps (and some other ways as well). But it has been impossible to get an aggregated and, importantly, an unduplicated view of the audience across all of these devices.

Currently, comScore is only releasing US audience data in the Multi Platform report. They are working on getting their global data into the Multi Platform format and I am even more excited to see those global numbers once they get the data cleaned up and QA'd.

This chart, from comScore's release on the Multi Platform service, tells the story well:

Media metrix multi platform

There's a bunch of interesting stuff in there. Take Google. They have 189mm users in the US on desktop/web. They also have 109mm users in the US on iOS and Android. But the unduplicated total audience is 211mm, meaning only 22mm of Google's users are mobile only in the US.

Pandora and Twitter stand out as highly mobile user bases. Pandora's mobile user base is >2x their desktop/web user base. Twitter's mobile user base is almost equal to their desktop/web user base.

If you are a Media Metrix user, you will see the option to get Multi Platform data on your key measures reports. I have been using this service in beta this past week and it is very useful in many ways.

comScore uses a combination of panel data and self reporting by websites and mobile apps to produce its data. The panel data is quite good for large mainstream properties. It is not as great for small websites and apps or services whose audiences are less mainstream and more niche. For those services, self reporting is a good idea and comScore allows you to self report for free by including their tags in your websites and apps.

I hope the entire analytics industry (first party and third party) follows comScore's lead and offers a multi-platform view as the standard view. That's the reality of how users access services today and the analytics industry needs to reflect it in their products.

Disclosure: I was a venture investor in comScore from the late 90s to the mid 00s and was on their board for almost a decade. I still own a few shares personally which I have no intention of selling.

#VC & Technology#Web/Tech

Comments (Archived):

  1. Emily Merkle

    Cool that you got to beta test…does Media Metrix handle performance-based advertising figures (like PPC)? That is an area they have been slow to attend to in a comprehensive way.

    1. fredwilson

      I am not sure Emily

  2. Kasi Viswanathan Agilandam

    interesting to find yahoo and AOL on the top-5….sounds like a decade old data :-)World wide view would be nice to watch… will tell more on who is gaining popularity in east and west and north and south of the globe.btw when you say ‘few’ … is that few as in few-thousands or few-mm ? 😉

    1. Anubhav Mehrotra

      Easy to ignore that these are ‘Properties’ and not websites. Yahoo! includes its search, shopping, Zillow real estate, Answers, mail etc. etc. AOL includes huffpost, mapquest, and everything else aol.

      1. Kasi Viswanathan Agilandam

        That is the funny part of it … when they are up on the chart still their bottom line does not say so.The number-to-monetization ratio is pathetic except for Google on the top-5.and that’s why i was surprised not just wit the numbers.

  3. JimHirshfield

    I’d love to see a column showing % on mobile and change over time.

  4. jason wright

    “Take Google. They have 220mm users in the US on desktop/web.”this is across all Google products?

    1. fredwilson

      I believe so

    2. fredwilson

      turns out i was using the wrong row for Google. the number is actually 189mm. i fixed the post

      1. jason wright

        I got my numbers wrong too.the classroom curse of being a ‘W’ – present but ignored.p.s. http://thisweekinstartups.c…your name came up in the conversation.

        1. fredwilson

          Back right corner for life!

  5. takingpitches

    Fred, are those the total Internet numbers instead of Google’s?

    1. fredwilson

      Did I screw that up?

      1. takingpitches

        Think so unless I don’t understand the table.

        1. Emily Merkle

          Those are GOOG’s monthly figures across all GOOG real estate.

          1. Anubhav Mehrotra

            The numbers stated are more or less correct. You can refer to the official press release here as well:

      2. Richard

        should be 211 and 188, and 25 mobile only (million)

      3. MissJennShaw

        Fred… correction needed: “220mm users in the US on desktop/web” is for total US, not Google

      4. RichardF

        yes but we get your point!

      5. Aaron Klein

        Crowdsourced math! Yet another AVC feature. 😉

    2. fredwilson

      i fixed the post. thanks for picking that up.

  6. takingpitches

    Seeing some of these names evokes the same reaction I had seeing the Twinkees news two weeks ago: those guys are still around!

    1. brian trautschold

      i had a similar double-take… searched for a date for the table… pretty incredible

    2. JimHirshfield

      long shelf-life here too

    3. ShanaC

      yup, and popular in the ad business, oddly enough (why doesn’t bluekai do stuff like this)

  7. John Revay

    “They also have 127mm users in the US on iOS and Android. But the unduplicated total audience is 235mm, meaning only 15mm of Google’s users are mobile only in the US.”I read this a few times and missed how you got to 15MM mobile, as i drafted this comment went back and looked at the chart and got to the number ( 235-220), at first glance I thought Desktop + mobile = total, it does not – re: the point of your post unduplication

    1. jason wright

      is 15mm mobile only correct?108mm are desktop/ web only?

      1. JimHirshfield

        I don’t think so. It’s just math, and as valid as the data is. Take total, subtract desktop, and you’re left with Mobile Only.

        1. jason wright

          “They have 220mm users in the US on desktop/web. They also have 127mm users in the US on iOS and Android. But the unduplicated total audience is 235mm, meaning only 15mm of Google’s users are mobile only in the US.””They have 127mm users in the US on iOS and Android. They also have 220mm users in the US on desktop/web.. But the unduplicated total audience is 235mm, meaning only 108mm of Google’s users are desktop/web only in the US.”

          1. JimHirshfield

            Your 2nd paragraph example/variation is a false calc.With this kind of unduplication calculation, you have to back out (subtract) the larger of the two sub-components from the unduplicated number, because when you subtract out the lower number (mobile in this case) you’re left with some dupes. IOW, your 108 number is mobile-only + desktop-only, I do believe.

          2. jason wright

            thanks Jim. I’ll ponder and brush up on my math 😉

  8. Richard

    Panel data has a unique meaning in data science. See Wikipedia. How is comscore defining it?

    1. ShanaC


  9. Matt A. Myers

    Can’t Google Analytics support this? Perhaps not by default, and that they don’t publish the aggregates. I imagine Google has access to a lot of the niche data though, from their free analytics offerings.

    1. ShanaC

      If you turn on anonymous data sharing with google, then sort of. Everyone I talked to says it is an extremely meh sample.

      1. Matt A. Myers

        Okay. Maybe Google will beef it up to compete with comScore now then.

        1. fredwilson

          google serves the publisher. it is their internal analytics. comScore serves the advertiser, it is third party analytics. big difference.

    2. Richard

      Why the providers (att , verison) aren’t in the analytics space is a mystery. They could use their cell tire data and provide deep, deep analysis.


        “Why the providers (att , verison) aren’t in the analytics space is a mystery.”.Because it doesn’t hold the value you think it does. If it did you’d see on TV “Fred Wilson worth 50 billion dollars is giving it all away.” Fred knows his analytics and he’s not making billions.

        1. Richard

          Hmm. You makes as much sense as a politician on capitol hill.


            First, those politicians have the whole of the country paying to keep them in their lifestyle! So, why do they need to make any sense?.Second, is there a part of my comment about the analytics space that you would like to discuss?.To elaborate… Selling analytics, the numbers to be specific because I think that’s what we’re talking about, must not be very valuable or else people would all be doing it. I think the value is in interpreting the numbers for a given business context. So, the numbers are of little value alone..There is one important thing to keep in mind. An organization holding extensive statistics about internet usage may find publishing those numbers less valuable than keeping them quiet and interpreting them for their own business..Selling an *explanation* of the numbers would be valuable in my mind.

        2. fredwilson


      2. Matt A. Myers

        Violation of privacy / terms?

  10. RichardF

    Who are the 24 million that manage not to use the GOOG?

    1. fredwilson

      duckduckgo users

  11. Tom Labus

    What’s the logic for ranking MSFT # 2 in the Media Metrix rank?

    1. fredwilson

      they have a shitload of traffic and users to their various online properties

      1. Tom Labus

        I was just amazed that anyone was saying something positive about them!

  12. Aaron Klein

    Fascinating numbers. If anyone is a skeptic of “mobile first” I’d encourage you to start carrying a tablet with mobile data. (Not tethered…I’m talking native, open it up and it’s instantly on and connected mobile data.) It’s been a game changer for me during the last few months.There were a number of services that I had no desire to do on mobile because of screen size. Now I’m dying to do more on this form factor. In fact, I just moved Riskalyze to Xero for accounting and one of the biggest draws were the mobile apps.I’m a big enough believer in this experience that I can see a lot of services going to mobile only. If you can ship four apps – phone and tablet versions for Android and iOS – and hit hundreds of millions of users…why build for web?I’m not quite there yet myself…and mobile has a lot of work to do to make sharing simple and easy in a way that the web makes simple…but I’m fascinated to watch what Apple and Google do to make their tablet experiences converge into “full computing experiences.”

    1. awaldstein

      Echoes my thoughts well.I do things with my phone. Buy, check, remind, share. They are always discrete acts.I work and converse on my laptop.Community is still not a mobile paradigm for me as yet.I’m going to buy a mini tablet to see if I have a bridge…

      1. Aaron Klein

        It’s getting closer. Still hard to write and the workflow is a little tough. But everything is elegant and simple. The design decisions in software drive a lot of simplicity.The downside is that some products still aren’t doing the hard work of delivering full power in the context of that simplicity.Evernote is amazingly good at that. I can’t think of a function missing from their mobile apps.


        “I do things with my phone.”.How has that impacted your profits?

        1. Aaron Klein

          Responding to customers from my phone has had a distinct impact on my profits.What world are you living in?


            The one where I don’t follow the crowd unless it is proven to impact profits or increase reveneues..Are you attributing that impact on time to response or cost of response or ?

    2. Elia Freedman

      I agree. istopped using a laptop, personally. I now carry an iPad and a desktop system. It is a little weird as I used a laptop exclusively from 1998 until 2011.

      1. Aaron Klein

        I haven’t gotten that far, but there are a lot of nights that I leave my MacBook Air on my desk and only take the iPad home.I still love long form writing on the MBA on planes, though… 🙂

        1. LE

          “that I leave my MacBook Air on my desk”If all your work is on your MacBook air (if that is your primary machine) I think it is a risk to carry around your primary machine anywhere. (And I’m assuming you encrypt the drive, you do that, right?) A machine that is carted around obviously has a much greater chance of getting stolen, getting damaged, of not working, than something that sits in your office static.I’m amazed at the people who operate off a laptop and the office and then cart that same laptop home and all over as their primary machine. There are certainly ways around the problem of things you are working on at the office needing those things on the road or at home. I don’t believe the solution is to cart around a laptop. I think it is inviting a problem.One of the things I quickly noticed about the ipad mini vs. a laptop is how easy it is to potentially “loose” a mini vs., say a laptop or a phone. A phone is something that can sit in your pocket and that is something you are simply used to keeping track of. A laptop is large and sits in a case and it’s hard to not remember that. But a ipad mini is small. It’s not a phone and it’s not a laptop. It’s easy to bring that to Thanksgiving, put it on your mom’s table and leave without it.

          1. Aaron Klein

            Given my market space, we take a lot of care on that issue. My MacBook Air is encrypted, everything is in the cloud, and all of my devices have password triggers after 1 minute of inactivity.And yes…I’ve already left the iPad behind once. One of the great advantages of always-on mobile data is that I was able to instantly find it on my phone, lock it down, make it ping and got a call from the person who found it within 30 seconds.Can’t do that with a MacBook Air, but I wish you could.

          2. LE

            A good idea if you are not doing this already is to use something like “super duper” to clone your drive. The cloned drive can be booted on another machine providing the ultimate ability to make sure that “it works”. If you are traveling you can take the cloned drive and if you have a laptop hardware failure (or theft) you can simply boot someone else’s machine from the drive. An external drive can be kept (encrypted of course) in the hotel safe. The cost is trivial. You can also cart the cloned drive as an “offsite” backup. Saves the problem of having to reconstruct a machine from scratch (through backups, or even instal and configure settings, passwords etc.). This is in addition of course, not instead of, any other thing you may be doing.

          3. Aaron Klein

            That’s an interesting idea. Ultimately, I’ve always looked at it this way: all my stuff is in the cloud. I encrypt my Mac so that if it falls into the wrong hands, it’s somewhere between tough and impossible to get to my data.Meanwhile, I get a new Mac, sign into Dropbox and Google, and all of my stuff comes pouring in without a hitch.


        I just don’t get it. What’s so different about a tablet from a laptop?

        1. Aaron Klein

          The constraint of 10″ screen and not having a mouse breeds much simpler user interfaces.Touch is a very natural user interface and works remarkably well when well-engineered (Google and Apple excel at this).And because these devices are packed with a standard set of sensors, app developers put them to some fascinating uses. Geolocated reminders are amazingly cool. Depositing checks into my bank account with a smartphone camera is saving me 40-60 minutes a week.The iPad needs to go a little farther to be full purpose for me. Grow up some of the software, give it access to a file system…and then combine it with a keyboard + screen for heavy duty work and you’re in business as a desktop replacement.


            How much has it increased your reveneus?

          2. LE

            ” you’re in business as a desktop replacement.”That’s cannibalization. They aren’t going to do that until someone else draws first blood. (Would you?).Did you ever notice how even though the mac runs a native multi user operating sytem (and you can have multiple logins) and is certainly powerful enough, you can’t take a Mac and have multiple monitors and keyboards attached and server, say, 5 users at the same time? It has the power to do that and has for many years.Back in the 80’s I had a multi user system (Unix System V) and had maybe 12 dumb terminals all hooked up (non graphic of course) at the office. The typical desktop Mac is way powerful enough to do the same thing with graphics. You could have one box in a cubicle area and serve multiple people at the same time. (Unless they were all processing video of course.)Why can’t you do this then? Because Apple isn’t in the business of giving companies a reason to only buy one physical machine.The products are released to maximize sales and profits. The ipad mini doesn’t have the retina display. They will release that later so that I end up buying another one. Features and benefits have to be held back otherwise there is no upgrade path.

          3. Aaron Klein

            I hear you. But I think Apple is a company quite willing to cannibalize themselves before they get cannibalized.The iPad Mini is a case in point. Will it cannibalize some iPad sales? Sure. But they’re making the market bigger.Will an iPad powerful enough to replace a desktop cannibalize some Mac sales? Sure. But Apple could own 20-30% of the “computer” business, compared to the single digit market share they have now.

        2. Elia Freedman

          Do you mean why a tablet instead of a laptop? I find it hard to explain. It’s an experience thing. I find it much more convenient, much easier to carry. It feels more personal. There is nothing between me and the screen. The interaction is cleaner. I want to do something I though it on the screen rather than an intermediary device. I’ve wanted a tablet since the 1990s and bought one of the horrible and early Windows XP Tablet Edition devices. The iPad (and Google Nexus for that matter) just nailed it.


            I find instant on, the idea of me being on all the time that is, to be intrusive. More and more I see people doing half-assed work in a haphazard approach. I get choppy useless responses to important questions. I get told “I sent you an email” when I didn’t get any email. Some of the most important context is lost during the two days between question and response..Let’s see we’re in a depression bigger than the great depression. The stock market is being bolstered by a QEternity future taxes spending of $40 billion a month. More than 23 million people out of work and I struggle daily to find people who can just form a complete correct sentence. Let alone hold focus on a topic long enough to actually get something done..I try not to be negative so I must say the ability to connect anywhere anytime is great. I just don’t find it *being* used properly..Last thought… Are these new *features* increasing the effectiveness and productivity of the user or are they just a portable TV?

          2. LE

            “More and more I see people doing half-assed work in a haphazard approach. I get choppy useless responses to important questions.”Agree with that. In a sense it’s shifting the burden of figuring out what someone wants on the receiver as opposed to the sender taking the time to write clearly what they want or mean or need.Same as has happened in the rest of society (with shopping). You used to have a salesman that was knowledgeable and understood their product at the store and who could explain and tell you what you needed to know. Now it’s all on shelves (POP) and it’s “every man for himself” where you end up buying things that don’t do what they are supposed to do. Used to be that you could call up a company and get a receptionist (before phone trees) that would route you to the correct place or answer a simple question. Now the burden is on the person needing something to figure out how to get that something. This of course has brought costs down not saying that there aren’t benefits (or being a luddite). But some people are just fanboys for progress thinking everything is great without considering the drawbacks. In particular the people that don’t have to deal with the aggravation of the systems and ways of doing business that they implement.

          3. LE

            I think it also depends on whether you are a good touch typist or not. For me (while I really like as mentioned the ipad mini) I simply can’t communicate as well without a good physical keyboard and large screen since the majority of what I do requires both screen real estate and typing.

    3. Richard

      I agree, but also have a sense that the story is going to be more complicated than the comscore data, with many factors dictating the access tool, age, profession, time of day, type of work, profession etc. I’m at dc 3rd wave coffee shop (filter) now and i see 100% of the users on a laptop, with about 20% using the phones. I’m on a laptop but listening to on my phone and twitter my iPad.

      1. Aaron Klein

        I still do tons of real work on my MacBook Air. It’s my tool of choice. But if I’m going somewhere to think, read and do a little writing, the iPad is winning my market share. I also used to carry my briefcase around like the nuclear football. Now the iPad goes with me…I’ve got a full browser and can respond to or fix anything anywhere I have a mobile signal.

        1. Richard

          What are your top 5 go to iPad apps?

          1. Aaron Klein

            EvernoteTweetBotKindleMailChromeAsana would be on that list if they had an iPad app!

          2. Alex Murphy

            Great list. You should consider two things Aaron. First, is a 7″ tablet. The size is perfect for reading, typing, holding in one hand and it can fit in your back pocket as you walk down the street. Second, try the Ted App. You can queue up videos to watch creating daily inspiration.

          3. Aaron Klein

            I seriously thought about going 7″. So far, I’m in love with the larger screen of the 10″ iPad and I’m getting more comfortable typing on it than I think I would on a 7″. I will have to try the Ted app! Thanks for the suggestion.

          4. LE

            The ipad mini is great. You need to get one. It makes the traditional ipad feel like a brick. And the 4lte is super fast as well. If you get the Verizon service you can tether at no extra charge.

          5. Aaron Klein

            I played with it and really considered it. Ultimately, I wanted to do more desktop replacement work with the iPad so I stuck with the 10″ version.Actually bought the iPad 3, and they released the mini + iPad 4 exactly 30 days later. Apple let me return the iPad 3 so I could have chosen mini, but I took the 4 instead.And yes, I did add it to my Verizon share plan for only $10/month. I replaced my $50/month MiFi device and now I’ve got mobile data for my MacBook Air with an eight hour battery (instead of the 2 hours on the MiFi).

      2. fredwilson

        i like listening to turntable on my Android too


      Aaron,That’s all last year’s news.

    5. fredwilson

      i am getting a Galaxy Note II. it is a hybrid between a tablet and a phone. i think its the perfect form factor for walking around with only one device

      1. Aaron Klein

        I’m interested to hear your take on it. They’re running some pretty good ads for it starring LeBron James.Of course…it looks like a regular phone in his massive hands. 🙂

    6. LE

      “I’m a big enough believer in this experience that I can see a lot of services going to mobile only. If you can ship four apps – phone and tablet versions for Android and iOS – and hit hundreds of millions of users…why build for web?”I think it’s important to learn, what I will call, the lesson of the SUV. People buy SUV’s because “it might snow” in certain cities where typically they only might need the SUV 1 day a year – if that. Same with many things people buy. People want to know that they can use something on a desktop if they want to do that. I think not building for the desktop is a mistake.

      1. Aaron Klein

        I don’t know anyone who doesn’t buy a smartphone because they might want to do something on their desktop.My point is: I think the day is coming when you can hit 100% of the market building for mobile devices.

  13. ShanaC

    I’m more shocked by the 15mm – does anyone have any idea if that number is going up or down (also, is tablet being defined as mobile?)

    1. Emily Merkle

      I imagine tablets are considered with mobile.What matters more than any other single factor is not unique visitors, but rather a combination of page views and ads per page.

      1. Carl Rahn Griffith

        People I talk to – with vested interests in device types being used/trends – seem to already be dismissing mobile meaning a phone by default – the emphasis is now on tablets.

  14. William Mougayar

    It is surprising that the Analytics industry wasn’t using multi-platform reporting to break-up Mobile/Desktop, etc.  Today’s self-reporting software is so good & detailed. We use Woopra & it provides an amazing level of real-time detail at many levels: user, platform, demographic, etc. I didn’t know comScore had insertable tags, and we will try that. We were using Quantcast. 

    1. Anubhav Mehrotra

      Woopra will not provide person based insights. These numbers are actual people. If the same person uses two browsers on a laptop – woopra treats it as two cookies and hence two people.

      1. William Mougayar

        You work for comScore, right? I think you should make it clear for everybody here. I wasn’t comparing Woopra to comScore as they are different products, so I’m surprised that you’re attacking that product. That said, Woopra identifies a user by a unique ID, whether they have 5 browsers open, they are seen as the same person.

        1. NatalieIssa

          I work for Woopra and William is correct, we track identified users by a unique ID. This means we track users across multiple browsers and devices (e.g. the user’s laptop, iPhone, iPad etc). Woopra’s focus on individuals is why we call the platform real-time customer analytics.

          1. Anubhav Mehrotra

            I am part of comScore’s Product Management team and the opinions here my own and do not necessarily reflect those of my employer. This was not an attack on Woopra by any means. I personally have used it and really love the platform. The differentiation was made merely on the basis of the two products being very different as William pointed out above.

  15. kenberger

    “I still own a few shares personally which I have no intention of selling.”– *ever*?

    1. JimHirshfield

      You buyin’?

    2. fredwilson

      i did not say “ever” but i am simply saying that they are set aside and i am not thinking of them as marketable

  16. Greg Bettineli

    According to this date, I would say LinkedIn might have more if a mobile problem than Facebook. Also, interesting that Amazon has the same mobile concentration as eBay despite what you might read or hear in the press.

    1. fredwilson

      i try to love linkedin’s mobile app but i don’t. i find it clunky.

  17. Judd Morgenstern

    Having a little trouble understanding the ‘unduplicated’ figure…EG following the math in this thread, Total Digital Pop – Desktop = Mobile Only Pop. So for Total audience that is approximately 235M – 220M = 15M unduplicated, mobile only users.However if you do the same for properties like Google, Yahoo! and FB you get 22M, 25M and 28M, respectively. Obviously their mobile only audience can’t be larger than the (15M) Total Mobile only population.If anyone from comScore is on this thread would love to hear how figures are calculated!

    1. Anubhav Mehrotra

      Hey Judd, At the top line level 15MM represent people that access the internet exclusively from a mobile device (smartphone or tablet). Now at the property (site) level, FB for example there might be additional people, that access the net from a PC but access FB ONLY from a mobile device. I personally know a lot of people that fall into this category. This in turn increases the ‘Exclusive Mobile Only’ audience for FB. Same concept applies to other web entities leading to a greater number of people accessing the internet from only their mobile devices at the entity level compared to the top line.


    “I hope the entire analytics industry (first party and third party) follows comScore’s lead and offers a multi-platform view as the standard view.”.Why? Internet access is a commodity now. Why is it important to know unless you’re selling access devices?.”I still own a few shares personally which I have no intention of selling.”.Just curious, are you getting a bunch of offers to buy your shares?

    1. Aaron Klein

      Oh goody, a new troll. Welcome to AVC.


        No, I’ve visited here bunches over the years, Aaron..I just view things in a simplistic manner. I also have worked in the computer technology field for 20 years. Internet access is a commodity now, even if you don’t see that..Also, it was strange that Fred made it a point to mention he was *surely* not selling his shares. So, I was wondering if that meant something I should look into..My motives are making money not clowning around!

        1. LE

          “which I have no intention of selling”I didn’t draw any conclusion but I found the phrasing interesting as well.

    2. fredwilson

      nope. i just wanted to be explicit about my comScore shares

  19. Abdallah Al-Hakim

    It is really interesting that the weather channel gets more visitors than Twitter, Yelp or Pandora !!

    1. fredwilson

      remember that this data is US only. i bet that on a global basis Twitter gets more.

      1. William Mougayar

        And this year, there was increased attention on the weather due to Sandy & other hurricanes, which Twitter traffic also benefits from.

      2. Abdallah Al-Hakim

        Noted but it is still impressive. This reminded me to download the WeatherHD2 app by vimov (iPhone) which I believe was at one point of of the top downloaded apps… – they might have one for android but I didn’t check.

      3. Dave W Baldwin

        This is probably associated with the time sElement issue I’ve mentioned, get the temp from weather channel on the fly vs. more extended browse on twitter.

  20. Chris Phenner

    Awesome that comScore is there insofar as multi-platform, de-duped audience reach, but as I craft/sell media plans to clients, the freely-available data available via FB/Twitter/SU dashboards makes ‘sampled reach’ feel quaint.And I’m thinking ahead to a day when FB offers an audience pixel to sites/apps that — when placed — gives them ‘real’ (not extrapolated) data based on authentic attributes entered via FB (especially) and Twitter (less so).I guess I’m seeing a third-party tool for ‘media planning’ getting usurped by DIY dashboards (which make APIs available), and with that, I’m seeing panel-based, extrapolated data being replaced by nominal data via social media’s authenticity.NOT trying to be a wet blanket; just reacting in the manner I’m seeing media planned.

    1. fredwilson

      as long as there are sites that don’t play ball with quantcast, facebook, or whomever, there will always be a need for a panel to supplement them

  21. Phil M

    I still don’t understand why comScore is de facto measuring stick for online traffic reporting. In my ten years of online publishing, I have never found their numbers to be accurate. It still really surprises me that the industry hasn’t moved to a direct reporting service like Quantcast, or that Google Analytics has never published their reporting data. When nearly every website has some kind of direct traffic reporting system installed, there has got to be a better solution than comScore… #startupidea?

    1. fredwilson

      comscore offers both direct reporting and panel based. the panel gives them data on everyone. quantcast’s reliance on self reporting means their numbers on those who don’t self report are awful.

    2. LE

      “I have never found their numbers to be accurate.”There is some kind of story or parable about a scale in a small town in which is calibrated incorrectly and as a results over charges everyone. Does it matter? You can only buy meat at one butcher and the butcher charges per lb. What’s the difference between $5 for a lb. and $4.80 for a lb.? In the case of these measuring services, the important thing is whether they are accurate relative to others they are ranking. While I don’t know the answer to that, I do know that in general even if you are the number 5 and your own internal data shows higher numbers that is the case for the number 3 and the number 150 as well.

    3. Dorian Benkoil

      I too am surprised Quantcast hasn’t gotten more traction. Show’s that a tech solution isn’t necessarily what it takes — marketing and share of voice also matter. Now that comScore has chosen to map cookies and panels, they can claim to give data that’s similar to Quantcast’s in statistical validity. It comes down largely to scale: who has the most direct measurement. If Quantcast tags aren’t on enough sites, Quantcast loses (as Fred I just saw, notes below. One service for whatever reason sometimes takes hold so it, at least, appears to be an apples-to-applies comparison for people looking at the properties. Everyone knows for TV, as an example in a different medium, that Nielsen isn’t “just so,” but they still use it because it’s agreed on as the de facto standard.PS All the services are inaccurate and disagree with teach other. Quantcast, Nielsen and comScore often fail to match. Google Analytics, using a different methodology, also diverges. The challenge is to understand the skew or prejudices of each and work with them in the context of a given property or properties.

  22. Techman

    If I remember correctly, I think that Disqus uses comScore along with other analytics.

  23. anand

    I’m surprised that AOL and the Ask network rank so highly. The last time I used Ask was when Jeeves was around 😉

  24. Dorian Benkoil

    Fred, it’s wonderful to start to get a picture of who’s consuming on what screens. It is, though, important to understand the limits and that the experience of how media are consumed can be quite different from the ways in which they are measured. I’ve responded at some length on MediaFlect In a nutshell: measurement chases consumption habits but will never quite get there as we increasingly personalize. There are challenges here which also spell opportunity for entrepreneurs.

  25. Nilan

    @fredwilson:disqus great post as alwaysLittle surprised though – you didn’t try to dig into what the source of the data was… this data means nothing without understanding a little more around the source in which it was captured.