Sentiment Analytics


Sentiment Analytics

Sentiment Analytics as it relates to speech, text and language – is the practice of attempting to understand the “emotional response” around a topic (person, place, thing, product, service, etc.).  I’ve been working in this area for over 10+ years.  My fist systems were interactive voice response systems where folks would leave a message.  We would use various tools to convert speech to text.  Language systems were very basic just a few years back.  Allot has changed.

Today, We use real-time pipelines to harvest data from Twitter, Facebook, News Outlets, etc. We use advanced libraries and now deep learning to understand what people Mean by what they Say.  Above is a sample of a product (dashboard from our pipeline) we’re developing ( Available Soon : Zapier, ifttt, Salesforce, etc. ).  The image shows a basic set of aggregations from the Blueskymetrics Sentimizer™ pipe-line and includes things like sentiment score, subjectivity, number of words used to convey meaning, frequency, etc.

There are many use cases for this type of analytics. The pipe-line itself is sophisticated to the point of offering realtime processing. ( See use case list below).  For now,  Check out my use case list for Sentiment Analytics.

Top Use Cases for Sentiment Analytics

  • Personalization [  if you don’t like Beer 🙁 , maybe you like Wine? 🙂 ]

Customizing a product and/or service to an individuals specific needs or preferences. The goals is increased customer satisfaction, improved website metrics and a better customer – brand relationship.  These metrics are strong indicators for customer metrics like LTV, MRPC, CAC and others. Contact US, If you would like to know how to build these important metrics. That’s what we do –

  • Customer Review – Feedback [ unhappy customer 🙁 bad product 🙁  – Churn Risk! ]

Customer Feedback, Review, Post, Tweet, etc. made by a customer who has purchased the product and/or service. There are many channels for feedback in modern systems including social media. There are side processes to understanding the sentiment that includes robot detection, ballot-box stuffing and competitive negative review identification.

  • Best Next Offer [ you don’t like Beer [bitter 🙁 ], maybe you will like a Apple Cider [ sweet 🙂 ]   ]

Fairly straight forward, Part of the personalization process.  We’re just trying to fulfill a need. Sometimes we may recommend a product with a competitor. “It’s better to make a customer than a sale”.  You may not make a sale; However, You can gain from the touch experience – lost sale analytics can help you with future business, supply chain and inventory planning.

  • Churn Analytics [  A huge area, you have the power as the customer ].

For most, Its accepted that the cost of acquiring a new customer is much higher than the cost of keeping an existing customer. The cost is so high, its a key metric – CAC : Cost of Acquisition.  There are many ways the calculate this cost; However, the point is that we want to minimize CAC and keep our existing customers.  Churn Rate is the measure of the number of customers that are moving out of a buying group over a measurable period.  New Customer, Churn Rate, CAC and others are key to understanding your business and whether you can sustain and compete in your market.

Do you want to compete like Amazon, Google with your Product and/or Service? You can – Contact US, We can help you build the right enterprise data warehouse and metrics to enable your business.

  • Increase Advertising ROI.
  • Lower your CAC.
  • Lower your Churn Rate.
  • Understand what customer are really saying about you and you products.


  • Customer Loyalty – Reward for Influence [  You have 1M+ followers, Yikes if you don’t like a product. ]

Get rewarded for your level of influence and a great review. Get rewarded for being a long-term customer that buys often and spends well.  Influence inventory levels, forecasting and product selection by being a part of the loyalty program. There is power here for the right customer.  These data sets are typically used for data mining to model behavior as a sample for the population.

  • Meta Message [ This is a tough area but allot of fun. Trying to understand what people are really saying.]

“I really love your dress, its so .. you .. and yellow  .. and the fabric. You  look great but its bothering me .. allot. To the point, that I will never buy that dress or anything that reminds me of the way you look. ”  Strong example. However, Yes – There are allot of times that reviews, merchandising and post on social media – backfire. We want to find potential problems before they happen and mitigate risk and avoid negative messaging. “Win because your better not because the other guy is worse.” -Gus

If you’re starting a new project or you need help with your existing data solution- Please Contact US.
Email: – Phone: 765.325.8373 ( call / text ).

Stay Informed: Please Contact US | Subscribe to our Blog Updates or If you would like more information.

Subscribe : Blueskymetrics Blog

* indicates required,  Managed By Mail-chimp – Please check your Spam Folder and Confirm Subscription.