Extracting value from voice of the customer data

In my last post, I talked about Voice of the Customer data and what it is.  The next question is, how do you gain valuable insights from Voice the Customer data?  The answer to that question depends on what type of Voice of the Customer data you are dealing with.

The unstructured Voice of the Customer data typically consists of voice recordings from call centers and text strings from emails, chats, tweets, agent notes, open ended survey questions, etc.  You could also consider behavioral data as Voice of the Customer data but that data is typically structured and thus easier to handle.  For this blog post we will restrict ourselves to the unstructured data.

Voice recordings require transcription.  Some customer interaction software like CallMiner have built in transcription.  There are also stand alone tools that will transcribe calls.  Transcription tools are not perfect and it is likely that cleaning and iterative processing will be needed. In addition, it is best if you can separate what the agent says from what the customer says.  Not all systems support that distinction.

Text scripts have the advantage of not needing transcription.  However, they have their own challenges as abbreviations and misspellings can interfere with analysis. Thus, these files will almost certainly need to be cleaned as well.

Once the Voice of the Customer data is cleaned and free form text, it can be analyzed.  Topic modeling will identify the reasons that customers are calling.  Beyond identifying product, process and customer experience problems which I mentioned in my last blog, this will enable you to identify coaching and training opportunities for agents.

Further, sentiment analysis will identify how customers are feeling.  Sentiment analysis can be difficult some situations, for example, using agent notes or call recordings where you cannot distinguish what the caller said from what the agent said.  Sentiment is valuable because you can use it to prioritize efforts and gauge the impact of problems and policy changes on customers.

Unstructured Voice of the Customer data requires more cleaning and processing than structured data but it is well worth the effort.

What is Voice of the Customer?

Voice of the Customer is a commonly heard phrase in many businesses.  You may be wondering, what is it?  Voice of the Customer refers to feedback provided by customers and that feedback can come through a variety of channels including inbound and outbound ones.

Inbound feedback is typically feedback that comes when a customer proactively contacts a company.  Think contact center calls, emails or chats for example.  The customer is reaching out and telling you about your products and/or processes.  This feedback is typically free form voice recordings or text documents.

Outbound feedback is feedback that the customer has provided in response to defined queries.  An example would be a market research survey.  The questions are set by the company and the feedback is structured.

Some people even consider behavioral data to be part of Voice of the Customer data.  If you stop using your online banking account or no longer shop at a favorite store, you are providing valuable feedback even though you haven’t directly told the bank or the retailer how you feel.  I read somewhere that for every 1 customer who complains, there are 26 who do not complain.  They simply leave without saying a word.

Voice of the Customer data is important because it provides valuable insight.  By knowing what your customers are saying, you can identify problems with your products, processes or customer experience as well as product enhancements and other innovations.  In addition, you can use this information to rebuild customer trust and retain customers.

How to benefit from text analytics

The artist, Anne Truitt, believed that ideas floated in the air, available to anyone for the taking.  I thought of this today when I read a recent Forbes article on text analytics.  It combined ideas from two of my recent posts.  First, in order to at least break even on text analytics, you need a plan.  The author is preaching to the choir.  See my post titled “Plan your dive.  Dive your plan”.

Second, in order to create a plan, you have to consider the benefits that text analytics will provide as well as the costs.  Not all text needs to be analyzed.  However, it can help you spot high-priority issues or customer defection.   In the example of preventing churn, you need to know how much a customer is worth.  As I mentioned in a post on text analytics, overlaying customer data to customer feedback will help you determine what action you take.  For example, you may choose not to retain some customers who threaten to leave because they are unprofitable.

In summary, text analytics can be a powerful tool when used in the right situation.  But first you have to determine if its worth applying text analytics.

Plan your dive. Dive your plan.

Every sport has its own lingo and scuba diving is no exception.  For example, you will hear one diver say to another, what’s the viz?  In other words, what is the visibility when you are under water. But what comes to my mind most often is “plan your dive, dive your plan”. My scuba instructor drilled that into my head.

Why do I think about this so often? It applies to almost every project I manage at work and probably the same is true for you. You need to create a plan for each project and then follow your plan!

When you start a project, you need to define it.  This includes creating objectives, outlining the scope of the project and incorporating feedback from the stakeholders.  Next you need to create the project plan which details how you will meet those objectives through discrete tasks starting from project kickoff until the final delivery.  Part of the plan will be a timeline with a cushion for the unexpected delays and problems that will inevitably come up.  In addition, your plan must include a list of deliverables and most importantly the KPIs that will tell you if you have been successful or not.  It is vital that you determine your success metrics at the start and that they are consistent with your objectives.

Once a plan has been developed, reviewed and agreed upon, you need to follow it.  So often interesting findings in the data try to lure me away from my objectives. It’s as if the data is populated by sirens calling to me as they did Odysseus, as he tried to return home to Ithaca. I don’t mean you should ignore interesting findings if they merit exploration but you have a choice.  You can revise your plan to reflect them or make exploration of these interesting findings part of a subsequent project or phase. Be sure to fulfill your original objectives before moving onto something else.

Listening to the voice of the customer

Customers want to tell others about their experiences with your products and services.  Are you listening?

With text analytics you can quickly convert their unstructured feedback into data that can be analyzed for insights whether they be customer complaints about service, information about product defects or praise for a great customer experience.  This information can be used to quickly address problems and learn more about your customers.

There are a variety of Text Analytics tools available.  They will automatically categorize the text.  However, you will want to tailor the algorithms to capture key words for your industry or region.  If you have ever visited the Boston area, you may have been surprised by the number of times that the word “wicked” is used in conversation.  “Wicked” in this area is used to mean very as well as bad or evil. Similarly, your industry might have lingo that is special like “BOGO” which means “buy one get one”.

Text Analytics tools will also enable you to extract key words or concepts.  For example, you might want to capture your brand or products and see what people are saying about them.  They will also determine if the sentiment is positive or negative.

Overlaying what your customers are saying with their attributes – their loyalty, lifetime value and experience with your brand – will enable you to understand and compare feedback across customer groups.  This will help you prioritize and tailor your response.  Combining customer data with customer feedback will provide a context for what your customers are saying.