Creating Customer Experience, Cognitive Computing an Enhancement or a Deterrent?

Yesterday I received a new copy of CRM Magazine by Destination CRM . There was a point of view story about cognitive computing called Cognitive Computing Energizes the Enterprise, by Oren Smilansky which I felt compelled to embellish in this post given my past experiences with this topic and CRM. Before I begin if you haven’t read this article and would like to view it, I would suggest you click on this link. Mr. Smilansky did a wonderful job outlining where this functionality is currently, and where it will be in the future. 

Let me set the stage first. (I will not mention any product or company names because I would like the reader to be totally impartial. There are many products in the market that claim they function as Artificial Intelligence or Cognitive Computing, I am sure depending on who you talk to there will be opinions of which ones are better.

Two days ago, I went to a Company’s website. I wanted to sign-up for a test of which, because of my status, I receive a 50% discount automatically. Needless to say, something was wrong with their systems, I was charged full price for the exam. My next step was to look for technical support on how I could rectify this situation. When I clicked on the help tab I was greeted by a Chatbot.

For those that are not familiar with the terminology, from Wikipedia, “A chatbot (also known as a talkbot, chatterbot, Bot, chatterbox, Artificial Conversational Entity) is a computer program which conducts a conversation via auditory or textual methods. Such programs are often designed to convincingly simulate how a human would behave as a conversational partner, thereby passing the Turing test. “   

I typed in my question, “I am not receiving my discount on xxxx test, how can I fix this situation?” A moment later I was presented with three documents to choose from, totally unrelated to my question.  The computer program matched key words in my question and returned the answers that were closest to the words I used. 

My next step was to type in “I want to talk to someone live.” Again, a return of three documents not related to my question. Finally, I typed a few inflammatory phrases at the Bot, still nothing happened. Totally frustrated I closed down the webpage and started searching the site myself, for a phone number, and eventually found it. 

I used this illustration for a number of reasons. If I was a real customer, this event might cause me to leave your company. A number of companies use this type of technology and believe this is cognitive computing when in reality this is nothing more than an algorithm that is parsing keywords and matching them to articles in a knowledgebase and returning what it considers reasonable responses. 

Let me repaint this story in terms of the work I am doing with CRM and Cognitive Computing. Since I was on a site where I had to log in, in order to get to where I was, I would expect the Chatbot to be able to look at my CRM record and greet me with something similar to Hello Steven. (A simple return of my first name from a CRM Record). 

When I typed in my question, I would have expected the system to look at my record in CRM to see if I had the credential active in order to actually receive this benefit.  At this point if the system was actually configured based on business processes and not just on returning documents I would have expected that a ticket would be created in their CRM system alerting the support center there was an issue. 

According to the article I mentioned above, “ The Aberdeen Group found that when companies incorporating cognitive technologies to support customer service interactions, they have seen an 81% improvement in employee engagement rates”. Unfortunately, they say nothing about customer satisfaction rates. 

My point, there are companies using Cognitive Technologies to defer costs of valuable people time to machine algorithms.  The feeling is 80% of the questions can be answered within 20% of the time. There is absolutely nothing wrong with that, but they should go further than just search and retrieve. 

Giving “WOW” types of customer experiences should be the goal with all companies. This will foster a personal experience and in turn garner loyalty. In my case, I am a captive audience so I am not going to leave this particular company, but in today’s world where there are options, and information flowing so freely, I would expect the focus should be on retaining customers through WOW customer engagement should be just as critical as Organizational efficiency. Actually, and organization should include the former in their definition of the latter.  

Back to my scenario. I look for an answer -> I am greeted personally -> I ask a question -> Information stored in one system is retrieved and analyzed -> The system understands there is an issue -> a ticket is created in the CRM system, and sent to the proper support queue -> the system could send me a few suggestions on how to correct this situation, and also reports a ticket was created, gives me the case number, and a phone number I can call to check on the incident. -> the system also sends the same information to me via email. -> The system asks me if there is anything else I might require -> I terminate the session. In the background, an email alert is sent to the support personnel responsible for these types of issues, and the entire chat transcript is saved in the CRM record, for support context and future analysis. 

This is a simplistic example, my argument is companies should look at their business processes from end to end and incorporate the flow of these processes into this technology, including a feedback loop. By thinking through your processes, you might be able to capture an additional 5% of the issues that are taking the extra effort to complete, without manual intervention.   Wouldn’t it be a hoot if the automated system could make a call to the core system through Business Process Automation and fix the discount switch for me? This isn’t science fiction, eventually we will get it right.  

Steven