How do you measure the number of people your digital marketing efforts reach? More specifically for those of you using content marketing, blogging or similar activities on your website, how are you quantifying that success? Is it Google Analytics or other website measurement tools? Is it the number of RSS feed readers for your syndicated content? A combination of both?

To put it bluntly, if your CEO passed you in the hall five minutes from now and asked, “How many people read our blog each day?” what number would you quote?

Google Analytics v2.0
Image by vrypan via Flickr

Erik Deckers, Paul Lorinczi and I have had an ongoing debate and discussion about measuring the impact of your content. While none of us are foolish enough to think you can definitively measure how many eyeballs consume every bit of your content (someone can always copy and paste it into an email that you never track), we all agree the current playing field of website analytics packages doesn’t do enough to quantify content marketing metrics.

(Erik and Paul run ProBlogService, by the way. I’ve used them for a couple of client content needs over the last year and have been quite happy with the results.)

While I’ve dabbled in explaining the nearly impossible task of determining how many people consume your content before, it just seems to me it’s high time someone figured this puzzle out and productized the solution. Since I stay fairly busy helping clients solve problems and most investors would laugh at the notion of letting me drive the bus, I’ll pour out the thoughts for all of you to run with.

The Problem With Website Analytics And Content

Let’s use Social Media Explorer as an example. Not because it’s all about me, but because it’s the one business I have total access to all analytics services for. Fair?

Google Analytics tells me that from Oct. 28, 2010 until Nov. 26, 2010, 55,401 unique visitors and 44,641 absolute unique visitors graced the hallowed halls of SocialMediaExplorer.com. They accounted for 70,210 page views.

Quantcast, a service whose code I also have installed directly on my site, and should thus also be an accurate measure of my website’s traffic, has different numbers. (They have different methodologies, as do most analytics services, which is the crux of the problem. But bear with me here, the numbers will illustrate the point.) According to them, Social Media Explorer saw 36,700 (rounded to the hundreds) “global people” in the exact same time frame. They also report 47,400 “visits” and 70,500 page views (okay … that one is close).

But keep in mind that for many of you, the content on Social Media Explorer is accessed via RSS feed. This means you never actually visit SME and, thus, do not count as a visit, unique visitor or page view. For that audience, we have to turn to FeedBurner, FeedBlitz or similar RSS feed management software. I happen to use FeedBurner (because I mistakenly thought when Google bought it in 2007, they would solve this problem).

Feedburner says from Oct. 28 until Nov. 26, an average of 16,043 people per day accessed my feed. That’s 481,290 people in a 30-day period. (Don’t worry, I called bullshit on it, too.) The problem is that many of these 16,043 people are the same people. My reach is not half a million. But that number is also inflated as it counts the number of times your feed is requested, not read. Feedburner says 338 subscribers accessed my feed via Bloglines, which is all but dead. The subscriptions still exist on a server somewhere but the likelihood someone is actually using them is slim to none.

Fortunately, FeedBurner also displays a “Reach” metric that it says is the total number of people who have taken action — viewed or clicked — on the content in your feed. My “Reach” metric for the period was 1,159 people per day, or 34,770 in the month.

So, if you’re counting, my absolute uniques (Google and 44,641) and my actual RSS actors (Feedburner and 34,770) means my total audience is 79,411 for the 30 day period. But wait! Is Quantcast’s 36,700 “people” more accurate than Google Analytics’s 44,641 absolute unique visitors?

And what about the social media monitoring and SEO aspects of your content? What other blogs have picked up and republished your content? Shouldn’t you also tally back links and SEO value for the content you’re providing?

Now I’m crankier than a gay rights activist pot head after Election Day.

Regardless of the coolness of the tool, the standards, measures and methodologies differ from service to service. I use Posterous for my personal blog. A whimsical list of 15 Things My Family Has Taught Me posted two Fridays ago 45 pageviews and 43 unique pageviews that day, according to Google Analytics. According to Posterous, the post was viewed over 1,000 times in the same time frame. While Posterous’s help topics point out that a post view is counted each time the content is loaded, whether on the blog’s home page, the post page itself or even when it’s pulled via RSS (which is a lot closer to the number I expect to see for a post’s metrics) and that Google Analytics is better at measuring visitors, there’s a big damn difference in 45 and 1,000+.

There are two primary takeaways from pointing all this out. Consider these and tell me what you think in the comments.

All Measures Are Guesstimates

Since there’s really no mark of consistency in all these measures of the number of people who consume your content, we are forced to pick one or two and monitor trends and percentages, rather than totals. We cannot say that 1,000 more people saw Friday’s post over Monday’s in definitive fashion. What we can say is that our numbers went up 6.5 percent.

Pick a couple of measurement platforms you’re comfortable with (ease of use, reports, exporting data, etc.) and stick to them. Look for trends and percentage changes, not hard numbers. If you must report a hard number, go with unique visitors in Google Analytics (or WebTrends, etc.) but explain to your reports that analytics isn’t perfect. The numbers you’re reporting are at least from a consistent source.

We Need Better Measures

Google owns Feedburner. I thought one of the reasons they bought the company was to marry the RSS service to its Analytics package and give us more complete reporting of what happens to and with our content once it’s published. While I’m not sure why they’ve never put two and two together here, (though they unveiled a new design beta for Feedburner last week and might be starting something) someone needs to build a better metrics mousetrap. The first company that does will get a huge jump on the competition. (But since the competition’s metrics are mostly free, they’ll have to have some compelling reasons we’d pay for it.)

And while I’m sure some community manager from Compete or similar service will want to jump in the comments and offer their $200 per month entry point services, they don’t do it either. Services like these use aggregated user data from Internet Service Providers, then extrapolate those numbers out over the total Internet population. It’s the Nielsen way of doing things. Ask 100 people, multiply their preferences by the U.S. Population divided by 100 and you know your statistics. While the ISP data is far more data than Nielsen’s, it’s still monkey math. At least Google and Quantcast measure stuff happening on your site directly.

Give me a service that I can plug in website analytics, content syndication analytics and maybe even backlink or social sharing data based on the individual pieces of content … and one whose measures are generally accepted as being a valid interpretation of the server’s logs or other data … and I’ll recommend it to the world.

If that animal exists, I’m guessing it’s tucked away in some $10K per month enterprise system that 99 percent of the world can’t get to. The question then becomes who will build the same mousetrap and set it free?

Okay stat geeks … tell me what I’m not seeing. Tell me what’s right or wrong with the existing measures. And tell me how you can combine those four major factors into a metrics solution I can just hit “Print” on. The comments are yours.

Enhanced by Zemanta
Did you enjoy this blog post? If so, then why not:Leave Comment Below | Subscribe To This Blog | Sign Up For Our Newsletter |

About Jason Falls

Jason Falls

Jason Falls is the founder and chief instigator for Social Media Explorer's blog. He is a leading thinker, speaker and strategist in the world of digital marketing and is co-author of two books, No Bullshit Social Media: The All-Business, No-Hype Guide To Social Media Marketing and The Rebel's Guide To Email Marketing. By day, he leads digital strategy for CafePress, one of the world's largest online retailers. His opinions are his, not necessarily theirs. Follow him on Twitter (@JasonFalls).

Other posts by

Comments & Reactions

Comments Policy

Comments on Social Media Explorer are open to anyone. However, I will remove any comment that is disrespectful and not in the spirit of intelligent discourse. You are welcome to leave links to content relevant to the conversation, but I reserve the right to remove it if I don't see the relevancy. Be nice, have fun. Fair?

  • http://scotduke.com mrbusinessgolf

    I have been asking the same questions for years and getting no answers that can be validated. Seems everyone has a tracking program but nobody can validate that the figures reported connect to a real live person. The only measurement I have been able to trust is accurate is the one I make when I walk through a gathering of friends and fellow social mediaist and ask if they read my blog..and even that I get from the cordial smiles that many of them did not read my blog but said they did to keep from being negative. Lot of work to find that the counts my analytics and stat reports are not close to accurate. Got to be a better way to prove to my business associates and sponsors that I am being read by more people than my stats show.

    • http://socialmediaexplorer.com JasonFalls

      Agreed, sir. Hopefully we can poke around and find some answers.

  • http://www.margieclayman.com Marjorie Clayman

    Yep – the questions get even more cranky-causing when you're talking about an advertising campaign. A few years ago, we placed some online space for a client and the site was telling us that we were getting monstrous amounts of clicks. “Yay!” We said.

    Our client's analytics program was showing that we were getting very very few clicks, and telling a client that someone else's analytics are better than theirs does not really cut the mustard.

    After arguing with the website for many many weeks, we did our own research, and the best answer I found was to get as many measurements as you want and look for trends and averages. All of the analytics programs we were looking at showed that the clicks went up this week, down that week, even though the numbers were different.

    So, if you look at all of the stats you quote above, you would still be able to tell the CEO with certainty “Well, we're averaging blah blah and all systems point to an increase in traffic comparing this month to last month.” Or something else.

    It's the best answer I've seen thus far. Though not perfect.

    • http://socialmediaexplorer.com JasonFalls

      Good additional info, Marjorie. Thanks for adding the thoughts. Awesome that

      in this infinitely trackable medium we can't figure out how to track stuff,

      isn't it? Heh.

  • http://socialmediaexplorer.com JasonFalls

    Good additional info, Marjorie. Thanks for adding the thoughts. Awesome that

    in this infinitely trackable medium we can't figure out how to track stuff,

    isn't it? Heh.

  • Kristen Findley

    Looking at specific numbers will drive you crazy – and get you nowhere. Web Analytics packages (all of them) are imperfect. Cookies, browsers, DNS, ISP's – all of the technology that enables our websites present challenges when trying to track behavior.

    Marjorie is right – look for trends and averages. I would also add ratios to that list. How many page views per visit, for example? And look for success events. What do you want people to DO on your site? Subscribe to RSS feed is probably one of those events – are those subscription numbers increasing? And what about signing up for emails? There's another success event. If you are feeling really adventurous, look at what pages people were visiting before they completed these events. Those pages contribute to your success events, so there's something valuable there!

    Finally, never compare numbers from one tool to another. They will never match – they all measure differently. Pick one set of numbers & watch the trends.

  • Now Seo

    Hi Jason,

    Great post and I'm looking forward to examining your question in my book on Social Media Analytics to be published next year by McGraw Hill.

    Since no standards in data collection and counting have been fully drafted or adopted yet, with the landscape so fragmented with different monitoring platforms and methodologies on one side, and different ways of understanding what it all means, there's probably no good way to answer this question today.

    Perhaps it's a discussion I'd like to be looped into as well – be happy your numbers are as high as they are…

    marshall

  • http://mytwittertoolbox.com David Perdew

    Oh, if there were such an animal. It can be really frustrating to read one metric that points upward and another that reads down, and the lack of a true standard has to be the top limitation of someone doing analytics today. I agree that we need better tools that talk to each other, and I hope someone comes up with such a breakthrough in the near future. It's hard sometimes to understand how much of an impact you're really making.

  • http://www.kherize5.com Suzanne Vara

    You always find that question that cannot be answered. The numbers are flawed, this we know and we are unable to accurately establish a metric to determine who is really consuming our content. Pretty scary given the technology before us. While I am not a numbers person as we can skew them to make them what we want them to be in certain situations, we do have to rely upon them in some sense to see the progress we are making.

    Clients want the numbers. Period. They need to match the numbers to the what they are paying for your services. If we showed them the above, they would be confused and want to know how many people are reading the blog and visiting the site. Hard count numbers.

    You give some great advice as to picking one that you are comfortable with and sticking to it.

    • http://socialmediaexplorer.com JasonFalls

      Thanks, Suzanne. Just trying to poke someone into fixing it! Heh.

  • http://socialthreat.com Davezilla

    Totally agree with you Jason and I'll add this. Google doesn't even agree with itself. Have five people all perform the same search on Google Blog Search at the same time and three of the five will usually have different results. Wait. This is for Google's own data! WTF? You can't win.

    • http://socialmediaexplorer.com JasonFalls

      Amen to that.

  • http://twitter.com/dakotareese dakotareese

    Like it or not, media will always be measured in terms of Reach & Engagement. Reading, or Acknowledgement/Comprehension, metrics are somewhere in-between those two.

    The trick, and I know several smart people working on this problem, is putting the right social spin on those older concepts.

    I'll promise you one thing though, the answer has nothing to do with Page Views.

    • http://socialmediaexplorer.com JasonFalls

      Don't disagree with you there. I've not used page views as a metric ever.

      (But to know for certain, I didn't have to report web analytics to anyone

      until about 2005 when pageviews began to be called to task for being

      worthless counts.)

  • http://www.terametric.com Taariq Lewis

    Hi Jason: Great post and you shared some great questions! It's late and super-cold in Boston so forgive me if my follow-up is a little jumpy.

    Conclusion: Web Measurement tools currently are mostly free because the value-add they deliver is very low, relative to the cost of using them. Markets for software tools are dependably efficient so you only a programmer with lots of times on their hand would generate yet another counter tool to use. As such, measurement must evolve from simple counting to combined count-analysis-action in order to be of any value to marketing folks. Given what you've requested above, I don't think you'll get a tool anytime soon that will be 100x better than GOOG or Feedburner analytics.

    So, what's the real objective of your measurement, Jason? If you knew the full reach of your content, what would you do with that information? How would it make you another $XMM in revenue or how should it help you secure another 50K of readers? How should metrics make your day 1000x better? I think marketers today are still hung up on answering the boss: “We got a 10% jump in traffic that helped contribute to a 20% increase in white paper downloads. Yoo hoo!” However, marketers don't think like CEOs or CFOs. As such, their metrics lack a strategic approach that folks in finance leverage far better.

    I think the question you ask is a good one, but I think its just scratching the surface of your business objectives. Let's talk about where you want to go, how you plan to get there and how metrics enable you to optimize your choices so that you work less and generate more wins or revenue or any other business objective.

    You were right. The most powerful metrics are sitting in bigger business intelligence systems and don't see the light of human eyes because they are part of a holistic approach to measurement which is anchored in modeling impacts and modeling activities to maximize return on invested capital and resources.

    If I had a measurement tool that could tell you how to increase your reach by 20% over 1 week so that you could achieve your business goals, would you buy it? Would it make your day easier? Would you smile?

    Cheers,
    Taariq
    http://www.terametric.com/blog

  • http://www.terametric.com Taariq Lewis

    Hi Jason: Great post and you shared some great questions! It's late and super-cold in Boston so forgive me if my follow-up is a little jumpy.

    Conclusion: Web Measurement tools currently are mostly free because the value-add they deliver is very low, relative to the cost of using them. Markets for software tools are dependably efficient so only a wealthy programmer with lots of time on their hands would generate yet another “web-counter” for us marketers to use.

    We need to evolve past simple counting to combined count-analysis-action systems in order to be of any value to marketing and the greater organization. Given what you've requested above, I don't think you'll get a new tool anytime soon that will be 100x better than GOOG or Feedburner analytics. Those are good enough to get you about 40% of what you think you need 80% of the time. No?

    So, what's the real objective of your measurement, Jason? If you knew the full reach of your content, what would you do with that information? How would it make you another $XMM in revenue or how should it help you secure another 50K of readers? How should metrics make your day 1000x better? I think marketers today are still hung up on answering the boss: “We got a 10% jump in traffic that helped contribute to a 20% increase in white paper downloads. Yoo hoo!” However, marketers don't think like CEOs or CFOs. As such, their metrics lack a strategic approach that folks in finance leverage far better.

    I think the question you ask is a good one, but I think its just scratching the surface of your business objectives. Let's talk about where you want to go, how you plan to get there and how metrics enable you to optimize your choices so that you work less and generate more wins or revenue or any other business objective.

    You were right. The most powerful metrics are sitting in bigger business intelligence systems and don't see the light of human eyes because they are part of a holistic approach to measurement which is anchored in modeling impacts and modeling activities to maximize return on invested capital and resources.

    If I had a measurement tool that could tell you how to increase your reach by 20% over 1 week so that you could achieve your business goals, would you buy it? Would it make your day easier? Would you smile?

    Cheers,
    Taariq
    http://www.terametric.com/blog

  • http://www.terametric.com/ Terametric Team

    Thank you Jason for being honest about how difficult and inexact this is for marketers. Rather then get mired in the collection and manual analysis of data (which is where we are today), marketers will need to look for tools that are like the ones used in the financial services industry for decades. The practice of pulling disparate data sets together to see the forest beyond the trees, allows analysts to evaluate securities individually and against their peers (a la marketing channels like Twitter, website, blogs, etc.) and allows portfolio managers to choose the best portfolio mix (a la marketing mix) that yields the best performance. This is performance-based analysis over time which continually adjusts based on the daily performance of this mix.

    Let's face it – managing large data sets is a relatively new task for marketers. You're right that the current set of tools that exist merely allow us to count. We've got to move beyond counting and looks for solutions that do the counting, the analyzing and the optimizing!

    Thanks for another great post, Jason!

    • http://socialmediaexplorer.com JasonFalls

      Good analogy. Thanks for the thoughts!

  • http://www.terametric.com/ Terametric Team

    Thank you Jason for being honest about how difficult and inexact this is for marketers. Rather then get mired in the collection and manual analysis of data (which is where we are today), marketers will need to look for tools that are like the ones used in the financial services industry for decades. The practice of pulling disparate data sets together to see the forest beyond the trees, allows analysts to evaluate securities individually and against their peers (a la marketing channels like Twitter, website, blogs, etc.) and allows portfolio managers to choose the best portfolio mix (a la marketing mix) that yields the best performance. This is performance-based analysis over time which continually adjusts based on the daily performance of this mix.

    Let's face it – managing large data sets is a relatively new task for marketers. You're right that the current set of tools that exist merely allow us to count. We've got to move beyond counting and looks for solutions that do the counting, the analyzing and the optimizing!

    Thanks for another great post, Jason!

  • http://dlvr.it/ Bill Flitter

    Jason, great discussion. Metrics and analytics is a beast. Finding the right mousetrap that fits everyone and “set it free” is even more challenging. Over the course of the last 5 years we've built analytics tools (Pheedo & Dlvr.it) for publishers. We listend to what they have to say, gave them the features they requested and then circled back. After all the pretty charts & graphs we created upon their request they focused on 2 things: Subscribers/Followers and how that translated into revenue. I would ask, “but what about x feature you had to have?” The response I got was – “haven't logged in to the reports for months.” We heard this over and over again. As we are building out Dlvr.it we focused on keeping things simple and actionable and are avoiding complex and complicated. I will leave complext up to Google or Webtrends – you get what you pay for, right? What is most important to you? If you care about your readers, you will care about what they liked and didn't like. What are they reading? What are they ignoring? If you are publisher that makes most of its money from advertising, you are going to care about the accepted stats package that media agencies 'believe' – Comscore or Omniture. My dream is stats package that focuses on a few key KPIs and allows me to adjust my business on the fly to meet those KPIs. Think about this way, how complicated in doubleclick? I see my impressions, clicks and tie that into my actions on the backend. I adjust my campaigns to where I am getting conversions (my KPI). Why can't content be that simple?

    • http://socialmediaexplorer.com JasonFalls

      Thanks Bill. I certainly agree with you that there's no magic bullet for

      everyone. We need analytics that allows there to be millions of magic

      bullets. I'm just optimistic enough to think it's possible to do that.

  • http://twitter.com/guy1067 guy stephens

    It's a really interesting question and thank you for sharing your insight with us.

    For me, it's less about the numbers and more about what you would do with them. If we accept that the numbers will always be flawed – flawed in the sense that we don't know how many people have truly done something or read something – how would that change what you want to do if we knew the numbers we got were absolute and real and not simply an indication?

    The numbers are what the numbers are, they aren't right or wrong. The measures are what the measures are. How we use them, what we do with them might be right or wrong in light of the numbers we are using. We create the disconnect not the numbers. So perhaps ultimately, are we sure of what we want to do with the numbers? In the end we often end up doing what we want to anyway, until we find a number that suits our needs.

  • http://borasky-research.net/2010/08/29/getting-started-with-the-social-media-analytics-research-toolkit/ znmeb

    I'm a statistician and I don't understand most web analytics software either. The answer is simple, though – if you don't understand something, it's useless and you shouldn't invest either time or money in it. Hire a human or two who understand statistics and who understand your business, give them access to your *raw* data and let them design a custom system for you. Give them a month and you'll be far better off than if you try to wallow in reams of counts and charts and dashboards and metrics and funnels from an expensive package.

    • http://socialmediaexplorer.com JasonFalls

      A strong bit of advice there. Thanks for the suggestion.

  • http://twitter.com/talgalili talgalili

    Hello there,
    I came by your blog through a twitter someone posted of it. How ironic to the topic…

    The only point I can add to this discussion is that you didn't mention action tracking systems like the one that google offers, or other nice heatmap services like clicktale.com or crazyegg.com

    The problem with webtracking falls between three domains:
    1) Integrating different data sources of consumer usage of the site.
    2) Wrongly detecting who is a person and who is a spambot
    3) Detecting the interaction of the reader with the content (in short and long term)
    All of these, together, leave room for a lot of innovation.

    Maybe I'll try to think about it more in one of my future posts… (http://www.r-statistics.com)

    Cheers,
    Tal

  • Walter Knapp

    Jason – good post, and I agree there doesn’t seem to be one version of the truth. But I think the value actually exists more in the comparative context of your site. Meaning that the trending, and other site to site comparisons are more useful. Visitors or other site-specific stats are points of time and therefore just a data point. The trick is to figure out what it means.