Borasky Research Journal

Richard Stacy asks, “Is much of social media monitoring snake oil – or have I missed something?” In his blog post, Stacy first describes how he sets up social media monitoring for a client, using his expertise in available tools and knowledge of the client’s business environment. He then goes on to say, “However, out there is a huge industry selling incredibly impressive black boxes that reel off reams of charts and data and figures and tracking, with sentiment analysis and conversation mining (conversation mining?) and all sorts of other wizardry.”

Stacy concludes with, “But there again – it appears to be big business. People are buying these remote sensing and analysis products. Is it just snake oil and are the purveyors of such trading (albeit unwittingly) on ignorance? Or have I missed something?” In my opinion, Stacy’s custom approach is clearly the way to go for a small business, assuming they can afford to hire a consultant like Stacy and have the staff and management commitment to keep the monitoring going.

But for a large business, with millions of customers and prospects, thousands of employees, terabytes of customer data and complex web presences, there’s simply no way the Stacy approach can scale. First of all, the free tools are difficult to work with. They require human intervention and constant tweaking. And in some cases, you need a developer who understands an application programming interface (API).

Second, the whole point of social media monitoring is to acquire and exploit clean and valid data about customers and prospects and how they interact with your products or services. You simply cannot build a such a clean database by hand – without high-powered statistical methods that require expertise in quantitative marketing and significant storage and computational hardware capacity.

But the real reason why the advanced monitoring and analytical tools are vital to large businesses is integration. What Stacy does by hand for a small business is integration. With the advanced tools, it’s possible to tie in-house customer and CRM data with sentiment, presence in the media, geographic and demographic data. And an integrated tool answers one other vital question: “How does the marketplace perceive your competition?” And it’s possible to leverage a small investment in software, hardware and analytical staff into a significant competitive advantage. You just can’t do that by hand.

So my answer to the question, “Is much of social media monitoring snake oil?” is, “Hell, No!” It’s a growing business for a good reason – it creates value by enabling businesses to read the “digital body language” of customers and prospects, and provides a competitive advantage at a vastly lower cost than other approaches.

Nobel prize laureate Herbert Simon once popularized the term “satisfice”. Simply put, “satisfice” means “make something good enough.” It’s what we do when optimizing — making something the best it can be — costs too much in terms of time, money or other resources.

I was reminded of this concept the other day while experimenting with Twitter search. In particular, I was attempting to figure out how the Twitter search “sentiment analysis” works. And the reason I was doing that is that I was struggling to understand why my screen name, “znmeb”, was returning a 6 percent “negative” sentiment rating.

So, why does a nice guy like me have a negative sentiment rating of 6 percent? As far as I can tell, It looks like Twitter search determines whether a tweet is positive, negative or neutral solely by the presence or absence of smileys! That is, if a tweet has a or or , it will be counted as a negative. If it has a it will be counted as
positive, etc. Incidentally, if it has a “?” it will be counted as a question. That, apparently, is the scope of its “sentiment analysis” capabilities.

OK … that’s part of the story … people who put frowny-faces in tweets about me will give me a negative rating. So … who are the folks frowning on yours truly? Well, in the words of Pogo, “We have met the enemy and he is us!” It turns out that most of the tweets counted were tweets I made! So, if I frown on Apple, Twitter search frowns on both Apple and me. If someone else frowns on AT&T and addresses it to “@znmeb”, Twitter search frowns on AT&T, me and the sender.

To test this, I posted a tweet that said, “I hate Hoover vacuum cleaners — they really suck ”. Now that’s an example of a sentence more or less deliberately designed to embarass any natural language processing sentiment analysis algorithm. As you probably have guessed, Twitter search counted it as a positive reference to both me and Hoover.

Why does this matter? Because it looks to me like many — perhaps most — of the social media monitoring tools available use Twitter search as their primary way of gathering input data. If these tools are passing on the sentiment analysis provided by Twitter search without any further processing, they may misrepresent the true sentiment about the search target. And that in turn may foster errors in decisions made using the dashboards these tools provide.

So … I put the question to my readers: “Is Twitter search good enough?”