Social media measurement: Sometimes a picture is worth a thousand tweets

In all of the posts I’ve read about social media measurement, very few address the possible role of qualitative research — measuring when you don’t have data — so let’s take a look at that today, shall we?  This will not be boring, I promise.

To make sure we’re all on a level playing field, let me quickly review the difference between QUANTITATIVE and QUALITATIVE data.

Quantitative marketing research is descriptive and conclusive.  It addresses research objectives through numerical measurement and statistical analysis.  In the social media world, this means data you can easily collect and measure like tweets, page views, comments, and perhaps even sales.  These are the facts and figures that get all the headlines. 

Qualitative Research is more, well …  touchy-feely.  It uses small samples and may involve focus groups, interviews, and behavioral observation.  Although it does not lend itself to statistical analysis* it can still be a quick and effective way to tell a story.

Because of all the free and voluminous data available through the social web, most of the attention is on the sexy quantitative side, but it might not be the best way to show value or tell your story.

Story time

Let me give an example from my own experience …

In addition to marketing and management, I also have a background in organizational development.  On one of my projects, I was delivering a training program to help correct dysfunctional management-union dynamics in a large company.  The people who went through the program raved about its effectiveness and had concrete examples of how it was dramatically improving the workplace.  The company’s top managers — who would not go through the program — were very skeptical about any progress and, lacking measurable results, were leaning toward cancelling it.  Like most managers, they demanded quantitative measurement … and I didn’t have it.  Sound familiar?

At the next employee training session, I mentioned that the program was probably going to be cancelled. The result was an out-pouring of outrage by both union and management participants. I had a video camera nearby for a training exercise and said, “Excuse me, but would you mind if I just turn this thing on to record your views?”

The group proceeded to tell story after story about the benefits of the training and also scolded upper management for not attending.  I edited the video to conform to the 5-minute executive attention span and played it during their next meeting. The managers sat dumbfounded and impressed as their employees passionately talked about the tangible benefits of the training. By the end of the meeting they all committed to attending the training themselves and expanding the program — without one pie chart!

Apply this to the social web

I use this example because like PR, marketing, or social media programs, training is very hard to quantify on a nice, neat spreadsheet.   This situation was a perfect time to use stories — qualitative data — to define value in a very different, yet compelling, way.

When you’re struggling to measure the value of social media marketing in your company don’t overlook the possibility of using qualitative stories from customers, employees and other stakeholders.  They might be showing up every day in comments, reviews, and customer meetings.

The technology of the social web offers unprecedented ways to capture and display this qualitative output.  And you know, sometimes all it takes is ONE story to provide more new insight than a dozen graphs!

What are your ideas?  What are some of the ways we can use stories to demonstrate the value of marketing through the social web?

*Michelle Chmielewski wrote in a {grow} comment that values can indeed be assigned to qualitative data to create numerical analysis. In effect this is how sentiment analysis is conducted. However, I was just trying to keep it simple today! : )

All posts

  • Pingback: uberVU - social comments()

  • This post has been needed for a while, and I’m a little ticked that I didn’t think of writing it first 🙂

    On the relative merits of qualitative measurement… I once did a qualitative study in an organization that was experiencing significant stakeholder backlash against their mission statement.

    Of course the simple solution was to change the mission statement, but the more complex issue was *why* the backlash was happening.

    The qualitative research I completed not only acknowledged how the stakeholders felt, but it also analyzed what claims the organization was making in it’s communication with stakeholders.

    In the end, I discovered far more about the possible causes of contention (that we could avoid in reforming the mission statement) than I could have through a series of quantitative measures.

    This, I think is another value of qualitative research: more clearly articulating why something occurs rather than speaking in terms of data correlation.

  • I think all three people, Mark, Michelle and Andrew, are essentially circling round one thesis:

    Quantitative research shows what’s happening.

    Qualitative research shows WHY it’s happening.

    A slight simplification, but highlights the importance of qualitative analysis.

  • Mark

    @ Andrew + @ Kimmo Your comments are spot-on. I actually taught an entire class on on marketing research in 2009 so believe me it was difficult to condense this important topic to blog-length! Thanks for contributing these observations.

  • Excellent point, Kimmo, I love that.

    Mark – great of you to continue the discussion, and I’m flattered you mentioned me, I noticed quite a few other commenters on your last post! 😀

    Michelle @Synthesio

  • zoe


    I just did a presentation titled “Death to Pie Charts” – I could have used that cartoon. I too talked about how pie charts make me think of pie. (if you are interested, here is a link to the presentation on slideshare –

    My career focus is also on qualitative analysis and I use infographs to tell the story. I created this one to show the story about the conversation on Twitter leading up to Podcamp Toronto 2010 –

    One point I have to disagree is your comment that qualitative measurement doesn’t include data – it does, it’s just that the format and organization is different than our numerical counterparts.



    zoë siskos
    Manager, Measurement Science

  • I love that you wrote this post Mark. No one is more guilty of focusing too much on numbers than I am. But you are right, the numbers leave out the why.

    I personally love to mix the two and provide a snapshot of what is happening, but then drill down to highlight exactly why coverage is spiking or not, why negative coverage is increasing or not, etc.

    I hope some posts examples of powerful stories! I’d love to hear more like yours! 🙂

  • Mark

    @ Zoe — Appreciate the opposing point of view, which of course is correct : ) Thanks so much for sharing the presentations. Great job!!

    @Rebecca — I am so numbers oriented too. I love to dig into the data. So it’s easy for me to overlook the power of stories and images and testimonies. This blog post was for me too! : )

    @Michelle — Thanks for stopping by today!

  • Haha, love the post and the title Mark. I am working with a local (20 shop) pizza franchise headquarters and they are trying to measure anything social and compare it with T.V. numbers.

    Any advice on how to sell the value, if the customer isn’t engaged online, and is skeptical of any “fad”.

    Props to Kimmo for the what and why distinction.

  • Mark

    @Brandon I’m trying to think of a way to answer your question succinctly but I can’t. I think it would be easier if we had a phone chat and I’ll try to help you. Bottom line, if the franchise thinks this is a fad (ie: waste of time) you will fail no matter what you do. DM me on Twitter with your email and we’ll set up a time to chat, OK?

  • Folks, if you click on Brandon’s name above and get a poor reputation warning, try instead of “”. Typos on the net may have really unexpected results…

  • Mark

    @Kimmo — Wow. Thanks!! Good save.

  • The last two days worth of posts have been great – insights and opinion on the topic of SMM are something I follow with great interest.

    Thanks for advancing the discussion Mark!


  • Pingback: Curated Stories Feb. 23, 2010()

  • Great post Mark! Really enjoying your blog and insights.

  • Oh my, no one likes pie charts anymore! Tough to take for a data-mining geek.

    In Social Media Monitoring software, there is a big emphasis on “analytics” (which is a word ruined by Google to now only mean “web traffic” – but that’s another story).

    The really smart data-miners (i.e. me) have a few tricks up our sleaves to turn the WHATS into the WHYS. We call it fuzzy-logic, double blind queries, and so on. Beyond the buzzwords, what it really means is to find trends, relationships, and correlations within all this massive data. You CAN filter out the noise, find the unfolding issues, and truly understand the big picture with all this tons of raw data. But it starts with the quanitative analysis first, then quality is revealed next.

    Long live Pie Charts!

  • Love it! When I did market research studies, we made sure to collect quant and qualitative data. Both are important. Like someone noted above, qual shows you why something is happening, quant shows you what happened. If you are using SM measurement, to take a pulse and make sense, it’s good to have both. One report we generate is called “topic discovery” – via an easy to digest tag cloud, we show most relevant phrases that occur alongside keywords. Not an exhaustive qualitative tool by any means, but it provides a starting point where you can start drilling in.

    At the end of the day, nothing can tell a story quite like, well, a story 🙂

  • Mark

    @ Frank — I would love to hear more about how you are distilling wisdom from the noise. This is a critical skill in the age of the information tsunamai! P.S. I like pie charts but have always been more of a bar chart guy myself.

    @Joseph + @ Chris Thanks for taking the time to tell me you enjoyed the article. Thank means a lot!

  • Mark

    @Maria It seems “easy to digest” and “tag cloud” are mutually exclusive terms : ) I usually hate those little buggers but I’m willing to keep an open mind! I’m absolutely fascinated by your idea here to tell a story through this format and would love to learn more.

  • Sorry to barge in again, but I just happened upon this Ad Age article on marketing measurement:

    To me, its main takeaway was the subtle distinction between “placing demands” and “continuous improvement”.

  • I really enjoyed this post, as I am huge fan of qualitative data (seems like there are many others out there like me too!)

    I think the problem with social media and qualitative data is the noise filtering. This will only get more complex as more social media conversations are blanketed advertisements or PR stunts. I too, would love hear more from Frank about how he separates good qualitative data from the noise.

    How do we know what comments are actually genuine in this medium? How can we start to get at the influence of these qualitative measures?

    There are so many more questions I could ask, but thank you so much for the post. I will absolutely be back!

  • Mark

    @Danny — I believe we may all be in luck. I’ve asked Frank if we would consider a guest post to flesh out his ideas and I think he may do it! Let’s all cheer to encourage him: Frank! Frank! Frank!

    Thanks for contributing to the {grow} community today, Danny.

  • Hurray for Frank, he can do it! Oh, wait, that’s me. 🙂 I can do it!

  • Mark this is fantastic, love the image of you whipping out the camcorder. Talk about the quintessential AGILE digital creative moment. Thinking on your feet here, capturing WOM buzz, and of course it paid off. Very inspirational for anyone, including teachers and trainers who value participant feedback and know it may move the needle with those who control budgets. Taking that video testimony “upstairs” was brilliant.
    I am left thinking that this case study you’ve shared illustrates perfectly the importance of us all investing some time to develop our digital skill-set so that we too feel as confident as you did to tape and edit a video spot on the fly.
    Great story and I will be sure to share it. Thanks.

  • Mark

    @ Frank Thou rocketh.

  • Mark

    @Sidney Eve As I recalled this event, you know what really took the guts? Showing it to management. This type of thing is just not done very often, at least in the blue-chip corporate environments I’m accustomed to. Our leaders are so conditioned to respond with their head based on numbers — how was I going to be perceived with this appeal to their hearts? At the time I thought there was a chance I could be tossed out of the room and my personal “brand” tainted. But the risk paid off and the experience helped me think more broadly about the use of story-telling to add depth to the data. Thank you for sharing your time with us today. It’s always an honor to have you comment.

  • Hey guys, while I’m writing up the guest post for Mark, you may enjoy an ezine article I wrote a few weeks ago. It somewhat fits this discussion. It’s about analyzing Tweets during a mini campus riot at UT (University of Tennessee). Have fun: (The link is to

  • I love this post. I think both types of data have their place. People do tend to focus on the quantitative data, and often disregard qualitative data. I have some training and experience in usability testing–and that data is primarily qualitative. You observe subjects using your site or product, videotape them, interview them, and see what problems they encounter or conceptions they have. At Microsoft, we always videotaped subjects, and often put together tapes with clips from the tests. Even the developers who were focused on quantitative data were usually swayed when they actually observed several people fumbling in the same area of a product, or heard similar comments.

    The entire discipline of usability testing is based primarily on qualitative data. It had an uphill battle getting respect, in part because of that, but now usability testing and its methods are widely accepted. Perhaps the same will be true of qualitative data for social media.

    Thanks for another great post, Mark!

  • Mark

    @Neicole — Superb example. Thank you!

  • intel_chris

    One thing I think is worth emphasizing is that whatever your analysis method is, qualitative or quantitative, in the end your data needs to tell a convincing story. My work is much more quantitative, but the end point of fashioning the convincing story is still relevant. Sure, you can bury someone in data or confuse them with rhetoric (the aphorism about what to do if you can’t dazzle them with your brilliance, comes to mind), but in the end if you really want to convince someone, you need a simple story and easy to understand evidence to back it up. I think the thing that made your camcorder example so cogent is that you had a story and you found a good way to illustrate it. Too often, too many people haven’t distilled their story to the key point and thus don’t have the clear example that demonstrates it. When you have that, you can trim your presentation down to the 5-minute attention span that you have available and your point comes though, and it doesn’t matter whether the evidence is a chart or a series of sound-bytes from interviews. The evidence relates to your story and that’s the take-away you want to get across in the first place.

  • Mark

    @Intel Chris — Really a superb point. The story has to be compelling and perhaps even entertaining — not a word normally associated with business presentations! Thanks so much for commenting Chris!

  • I am really late for this party, but will chime in anyway. It is possible to quantify qualitative data and convert it into very meaningful structured information that can be charted, but more importantly serve as comparison base for differentiation. We are doing it with measuring delta between customer expectation and experience in a pretty consistent and granular way by applying text analytics to customer reviews.

  • Pingback: Sourcing3 Buyer & Supplier Magazine - Top 24 B2B Marketing Posts in February()

  • Pingback: Social media measurement: Sometimes a picture is worth a thousand tweets « The *Official AndreasCY*()

  • Pingback: Social media measurement: Sometimes a picture is worth a thousand tweets | Social Media Butterflies |

The Marketing Companion Podcast

Why not tune into the world’s most entertaining marketing podcast that I co-host with Tom Webster.

View details

Let's plot a strategy together

Want to solve big marketing problems for a little bit of money? Sign up for an hour of Mark’s time and put your business on the fast-track.

View details


Send this to a friend