TwitterThere is no question that Text Analytics has come a long way in the last 3 or 4 years, and that the solutions available in the marketplace are solving real world problems. Unfortunately, our market is still dominated by too much flashy marketing and not enough down in the dirt numbers for “apples to apples” comparisons. Given that spring is coming and with it the annual Text Analytics show that draws all of the major vendors, I see an interesting opportunity for somebody like a Seth Grimes to orchestrate a test scenario for all of us vendors that would give everybody out there a real apples to apples comparison of the different technologies out there. I have no illusions that such a comparison test will happen, because of the inherant risk in not controlling the communication with prospects, but I can dream. Allowing my illusion to run a bit further… Seth Grimes strikes me as the ideal person to orchestrate such a bake-off because he isn’t closely affiliated with any of the vendors and has a strong desire to stay vendor neutral. Such a test would have to be of Text Analytics features like entity extraction, classification and possibly sentiment. All of these are features that can be judged by humans to create an authoritative set of results to measure against. Let me close this little dream with an open call to Attensity and Clarabridge (the two most well known players in this space, with technology I respect). What do you say guys, don’t the users deserve real numbers on how our technologies stack up?
Every once in a while, it is interesting to take a look back at how technologies evolve from the mundane to the sublime. Technology, in and of itself, is rarely the sole component of the solution to a problem. It is how people use it that helps to define its worth. The various online media outlets that we have been following at Lexalytics are no different. When we started working with content acquisition, web spidering was the primary avenue to get at interesting content. If you were a business, you had a website, your partners had websites, and your competitors had websites. And there was real business value in being able to keep an automated eye on your competitor’s websites to see if anything changed or anything new appeared. Our cell phones were also a lot bigger back then. Then newsfeeds hit the scene. Even better than webpages, here we had outlets of focused content that would update on a periodic basis. As a business, you wanted to watch the news feeds relevant to your industry, keep an ear to the ground to listen for which direction the buffalo were running. And cell phones got smaller. It wasn’t too much of a leap from newsfeeds to blogs. But, as I recall, the first time I looked at a blog it was on a friend’s website and the main thread of the discussion was what he was going to have for lunch. Another comment I’ve heard from the beginnings of blogs, “Blogs were just a bunch of message forums.” The technology was in place, but where was the application of the technology (where was the ROI). Over time, blogs have come to feature content that is more and more relevant to business. Business themselves are now finding blogs as a mechanism for a more dynamic dialog with customers and the general public. I first came across Twitter about a year ago. And again, at the time, it was not in a business context. An article in the Wall Street Journal at the time chronicled the content of the service at that time. But as with others that have come before it, business is starting to pay attention to Twitter. Why? First, a tweet about waiting on the phone with customer service, could easily be a tweet about your company’s customer service. That tweet about a new cool product, might be your company’s product. Twitter is becoming an outlet for customer opinions and discussions. And not only in the tweets that occur on the service itself, but as a gateway to websites (and blogs!) that are linked in tweets. Social media experts such as Chris Brogan are counseling businesses on why services like Twitter matter and how they can use Twitter to connect with their customers. We’re including mechanisms to gather from Twitter and analyze the content we find for our customers, to help them tap into and understand this additional source of information. Social media is evolving, it is finding a place in business, and finding a place to make an impact. PS. My cell phone has gotten bigger again, but now I can access Twitter on it.
I’ve been following a lot of conversations lately about ways to quantify social media marketing and ways to move the practice up the chain-of-command at an organization. Social Media is a rather complicated little practice that involves sharing a thought, via the written word (or maybe video/audio), over a variety of channels with an endless audience - so it got me wondering, too, how do you figure out the ROI on something so broad and so unpredictable? Since our business lives and thrives on the “voice” of the written word - and the more of it the better - it’s difficult to NOT put a number or figure on the value of what we can extract for your business using our technology. What would be the value for you to know the most frequent themes or concepts associated with your business or product? Or to figure out if you are perceived positively or negatively by the voice of your customers? Maybe putting value on that stuff is the easy part, but how do you know if that is an indication of a good or bad social media plan? Amber at Brass Tack Thinking posted a great blog that I think helps to highlight some of the fear businesses have with jumping into the analysis and evaluation phase of social media. She writes: “I have news for you. We aren’t going to figure all of this out neatly in time to take action without risk of failure. In fact, the time to take action is now. It’s yesterday. If we’re waiting for a tidy definition for ROI before we’re comfortable taking steps in a new direction, the train will have left the proverbial station. Have we always known and carefully calculated the potential success of something - especially in communications - before we do it? Do we ever have the capacity in a business sense to take something on faith, or at the very least on an educated guess? Can barriers be broken if we wait until all the details are neatly in a row? Moreover, ROI is something that you calculate after the investment has been made. It’s looking back at your efforts, and determining what you got out of them. The definition of “got” can be any number of things. Why are we so hell bent on making it fit one specific idea of what works? And can we not be content to do our best to plan ahead, but be bold enough to move forward and DO? Not having all the answers is not an excuse to be inert.” Obviously, we are always happy to discuss how Lexalytics’ solutions can help with that process of “looking back at your efforts” and ask, “What do you think would be the best way to calculate ROI of social media monitoring and analysis??” The more people thinking and talking about it, the better chance we have that we can put some metrics and measurements in place to quantify all that qualitative information.
Jeff recently spoke with Mike Vizard at eWeek to explore the importance of monitoring social media and how you need to look at the whole picture in order to get the best results. Have a listen to the podcast: Analyzing Social Networks
At the risk of kicking the horse just one more time, we’ve decided to take a closer look at the #Motrinmom’s controversy.Was the outcry really so overwhelmingly negative that the company had no choice but to cave to #Motrinmom’s Twitter tsunami and pull the ad? We wanted to find out. Lexalytics measured the sentiment (positive, negative and neutral) of a sample set of #Motrinmoms tweets from the weekend. We were surprised to find the numbers don’t necessarily support the notion of a “landslide” victory for #Motrinmoms. If Motrin had been able to see that, they might not have so quickly scuttled the campaign. To find out what really happened, we used the Lexalytics Acquisition Engine to retrieve tweets including #Motrinmoms through the open Twitter API. The API only allows for 1500 tweets so our analysis was limited by that number. We then ran those numbers through our Salience product to extract entities which included quotes about #Motrinmoms, twitter names, places, companies, etc. From there we could see who had more than 1% of the tweets, which we plotted as being positive or negative. This was a feel for influencers. From that information we determined if tweets were more negative or neutral/positive in regard to #MotrinMoms and saw that if you pulled out all of the neutral tweets (those with a sentiment of zero) we ended up with 58% negative tweets and 42% positive tweets. But whereas most campaigns are measured as neutral/positive vs. negative, that overwhelmingly shows that this session would have been 65% positive/neutral vs 35% negative:
- Thumbnail of sentiment per tweets
- Click for full-size image
- Thumbnail of sentiment distribution
- Click for full-size image
Twitter Results While this information can’t be applied to the whole #Motrinmoms tweeting universe, it does show how Twitterers who cared to use the hash tags in their tweets were trending on the company. The results beg an important question: was this was so overwhelmingly negative for J&J and Motrin?? We don’t know how much the ad campaign cost, but being national in scope we’ll assume there was some decent budget behind it. If Motrin’s brand managers were not just listening to the market, but accurately measuring it too they might not have been so quick to panic and pull the ad. But that decision would have to be weighed against a number of internal factors and it’s not our call. Maybe they’d pull the ad just as quickly (or faster); or, maybe they’d see the dust up as a great opportunity for more visibility (there certainly has been a lot of press about this), left it up but decided to reach out directly to the #Motrinmoms community. Of course brands need to be listening to their communities. But within those communities are individuals and not everyone within a community feels exactly the same. The ability to extract meaning from all the noise—especially in a crisis situation—and measure extremes of opinion might make a critical difference in important business decisions. Armed with information about what was inside the #Motrinmoms Twitter tsunami (not just knowing there was one) could have ended Motrin’s story differently and saved the company a huge headache (sorry). =================================== Ad Age offers a nice look at the Motrin-gate Social Media Case Study here.
Friday night was relatively quiet in our house so I logged onto my personal Twitter account to see how my fellow “Moms” were doing - it’s my night job. During the day I’m a “Social Media” tweeter. I just haven’t had the guts to let the two merge yet. But, because our company is in the social media industry I was fascinated by the #MotrinMoms phenomenon that was unfolding before my eyes. Mom, after Mom, after Mom was blasting an ad campaign on the Motrin web site. There were not millions of us tweeting about the “offensive” Motrin ad, and it wasn’t something that happened the moment it was released - it had been running since Sept 30 - but *just* enough of the voices started to emerge and share and blog and suddenly, *poof*. From an Ad Age article: He said the “Motrin moms” phenomenon wasn’t just the work of Twitter celebrities and marketing gurus such as Seth Godin, who also weighed in on the kerfuffle. “You don’t have to have thousands of followers to start something like this,” said Mr. Armano, who also blogs for AdAge.com. “Many people with small networks have just as much influence as a few people with large networks.” Goes to show - you can listen all you want, but you also need to find the right voices out there that matter to your business!
Jeff and I were talking recently about the voice of the cutomer. We love VOC, as it’s known…unstructured data is right up our alley.
What we were discussing was the importance of listening to your customers versus finding your customers. We all know the squeaky wheel gets greased, but isn’t there a lot more to your customer service than listening to the loudest ones? Isn’t it good to find your satisfied customers, too? Or the customer whose influence may not reach all of the country, but just one region, or territory, or radius?? People love to tell their stories, and a finding the ones spreading the positive is equally important as finding the negative.
In a time of economic downturn, retention becomes as important as growth. If you don’t reach out to all of your customers and find their stories, then you could be looking at a silent killer to your business - the semi-satisfied-but-not-overly-vocal-yet-totally-retainable-customer, that you let slip away.
Finding them is as good as listening to them. What do you think? Do you have what you need to know about your customers?
Since our release of Salience 4.0, there have been a couple of discussions about entity-level sentiment, and what it means. So I decided to write a posting based on a chat I had with our Chief Scientist, Mike Marshall on the subject. Let’s start with the basic method by which a sentiment score is assigned to an entity. The sentiment score for an entity is based on an aggregation of the sentiment scores for phrases that are associated with that entity. Observation #1: Sentiment for a thing is audience-specific and content-specific Think of recent news articles regarding executive compensations for companies that have been failing on Wall Street, or profits reported by oil companies. A record profit for an oil company would normally be considered a positive statement. However, when measured against the backdrop of other content indicating market concerns over the price of crude oil or consumer concerns of the price of gasoline at the pump, these phrases may weigh down the positive sentiment of the earnings report, and may result in a negative sentiment for a company that reports a record profit. Observation #2: The power of entity sentiment is in aggregation of results If you are a stock holder, you might not agree that record profits are a bad thing. And another piece of content that focuses solely on the earnings report may rate the same company positively. To get a true measure of the sentiment for an entity, you need as many data points as possible. This should smooth out individual results to provide a reasonable score for an entity. Observation #3: Machines can achieve the same result as humans on repetitive tasks, and perform them faster than humans In general, multiple humans judging the sentiment will agree with each other approximately 80% to 85% of the time. Machine-generated results will agree with human-generated results at approximately the same rate. The difference, however, is the amount of content a machine can process in a given span of time, and the consistency with which the machine will make its judgments. This also supports Observation #2, allowing many data points to produce quality sentiment scores for extracted entities.
Jeff recently completed a videocast with the great guys over at NetworkWorld.TV He explored why and how enterprise search and text analytics work well together. We often field questions about how to integrate text analytics into organizations and how to improve the results provided by enterprise search solutions. This should help with some of those questions. NetworkWorld.TV - Better enterprise search through text analytics
Now that the election has ended and the votes have been counted (for most, anyway - Senate runoff in Georgia?), we will be retiring our politicaltrends.info site. Don’t worry. We plan on being around in 4 years so we may fire her back up again, then. The information that came out of analyzing the political blogs was pretty amazing. We started with several candidates back in late 2007 and as they dropped out of the race, or got pushed aside, we watched coverage shift, hot themes moved from the Iraq War to Oil Prices to Financial Crisis. Hillary Clinton always seemed to be in the lead for negative sentiment, but it meant a lot more coverage and exposure for her early on. And while we never claimed to be political experts, we were thankful for an exciting and historic presidential race this year - it wasn’t just the choice in candidates that were “firsts” but the way we all communicated during this election. It certainly was a great showcase for our technology. It seemed EVERYONE had an opinion about the candidates. Thanks bloggers for being so passionate during this race. We’re confident our next venture of analyzing sentiment of Corporate CEOs on ExecDex will prove to be just as interesting. The blogging community has a lot to say and we love listening!
