Tips for Measuring Sentiment Analysis: Don’t Let Early Adoption Become Buyer’s Remorse


by Sandy Malloy, Information Services

A recent  blog post by measurement expert Katie Paine calls sentiment analysis the “latest shiny new measurement toy” and suggests questions to ask before attempting to measure  social media sentiment of corporate and other messaging.

One of her most important points  has relevance for press releases:  Do people express any sentiment at all in discussing your brand? 

Simple searches on Twitter and/or Google blogs will give you an idea, but first consider the kind of news your company announces.   If you sell measuring equipment to scientists, for example, it’s unlikely that sentiment will register anything but “neutral”–even if your products are discussed in social media.

No sentiment = no need for sentiment analysis.

That said, monitoring all types of media for what’s being said about your company, industry and competitors is important.  But monitoring is not analysis, and analysis might not be sentiment metrics–thus, paying for analytics when all you need is monitoring makes no sense.

Just for fun, I reviewed about 200 social media clips that a long-time subscriber to Business Wire’s NewsTrak Clips had received over a few days.   Almost all the tweets and mentions of  this engineering company were neutral.  One item stated that the company had been “called out” for some action. 

A human reader could easily identify the statement as negative, while software might or might not make that determination.  Either way, though, I would have had no trouble doing my own analysis with such a low volume of posts.  For Paine, the cut-off before investing in analysis software is 10,000 relevant non-spam mentions per month.

Of course, you can analyze your social media mentions in other ways to put this information into a meaningful context.  Almost every week, I see lists like this one that offer help managing and analyzing social media clips.

If you still can’t resist the “shiny new toy,” here are some steps to take to assure your early adoption doesn’t become buyer’s remorse:

1Start with monitoring and worry about analysis later.  You can use free tools or a more comprehensive paid service, as long as you can monitor how many posts or tweets come through daily.

2.  Commit to either looking at the results yourself or making sure a colleague does every day.  That way, the task won’t be overwhelming unless the volume is overwhelming –and then you’ll have a case for investigating analysis software after all!

3. If sentiment analysis is indeed a “key metric,” decide on whether the volume of posts and tweets justifies buying software.  Also, be aware that sentiment analysis software has improved but is unlikely to be 100% accurate.  You’ll probably need  people to cross-check at least a sample of the results.

4.  If you determine that sentiment is largely absent, decide what metrics are relevant and find the tool(s) that will help you gather the information.

One Response to Tips for Measuring Sentiment Analysis: Don’t Let Early Adoption Become Buyer’s Remorse

  1. […] the misguided emphasis on data that is debateably important (if not completely unnecessary, as I commented  recently.)   Just because software can measure a particular parameter does not mean it’s […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

Join 46,791 other followers

%d bloggers like this: