Loading...
You are here:  Home  >  Data Education  >  BI / Data Science News, Articles, & Education  >  BI / Data Science Blogs  >  Current Article

Chatbots Will be the Next Evolutionary Step in Business Analytics

By   /  April 10, 2017  /  No Comments

Click to learn more about author Ilan Hertz.

One day, probably in the not too distant feature, you will be able to talk with and put your questions to your Business Analytics system, as though it were a person.

Technology has already advanced significantly in this area with the arrival of chatbots, also known as artificial conversational entities. Within certain limits, these software applications are designed to engage human beings in interesting, meaningful conversation. In Business Analytics, they can allow you to interact with your business data using normal, natural language, rather than imposing a formal computer programming language that requires specialist IT skills.

Leveraging AI and NLP

There are two principal types of chatbots. One uses predefined rules, scanning for keywords in a sentence or question from a human being, to then search a database for a reply that best corresponds to the trigger words. The other type, more advanced, uses Artificial Intelligence (AI) to understand language and the knowledge it contains (“natural language processing” or NLP), instead of mechanically reacting to keywords. Some advanced chatbots can also become even smarter by learning from their conversations with people.

Simpler is Smarter

When chatbots (or simply “bots”) are linked to a suitable Business Analytics engine, capable of preparing data for analysis, analyzing it, and extracting insights from it, there are two big advantages for you as an end-user. First, it is easier to ask for the exact business information you want. Second, it is also easier to make sense of the answers, which also come back to you in the ordinary, natural language that you use every day.

Avoiding the Wastage of Context Switching

If you need specific and important pieces of information on demand and at any time, natural language query and immediate response can save you time and temper. Otherwise, productivity takes a hit each time you are obliged to open a dedicated dashboard application and crunch data before receiving answers. Context and task switching consume a disproportionate amount of energy, and are best minimized. Situations that oblige you to continually move from one system interface to another to retrieve information, transfer information, or perform specific calculations should be avoided.

The Chatbot is Your Friend

For example, if you wanted to check your overall sales from a given dashboard, you might simply tell your smart bot “Summarize Sales dashboard”. You decide what your bot can access. With your authorization, it can make use of all kinds of data sources, system logs and measurements, accounts, and reports. Your bot would then automatically find the data, extract the insights of interest, and send you the answer. You would not have to locate the dashboard by yourself, make calculations, or select the right visualization, to get what you need. Better still, a chatbot can include handy widgets, images, and charts in its response to help you consume your data even more effectively, much like a friend or helpful colleague would reply to your question.

Fitting in with New Work Patterns

A smart bot can also link to popular messaging, communication, and collaboration platforms such as Facebook, Skype, Slack, and Trello. This connectivity matches a shift in end-user preferences from traditional apps with structured interfaces, to chat-based communication and natural language processing. Younger generations, millennials for instance, tend to use rapid, text communications via mobile and messaging, and look for solutions at work – such as smart Business Analytics bots – to allow them to do this.

Integration with Daily Decision Making

Your bot becomes an extra communication channel, an additional conversation, all in natural language. Just as you can stay within your current conversation window in a messaging or collaboration app to exchange instant messages with colleagues and teams, and share files and other resources, you can also obtain the key business insights you want via the same window and in real time. Your bot lets you seamlessly integrate your data, big data, Business Analytics, and business intelligence into your everyday decision making.

“Bring Your Own App” is a Snap

Bots also fit into another growing trend, that of “Bring Your Own App”. In BYOA, a team within a business chooses its own messaging or collaboration app or tool, to let team members work the way they want. The right Business Analytics bot then provides easy, text-based, natural language question and answer facilities integrated into the team’s favorite app. The Sisense BI bot takes user-friendliness and productivity even further, by enabling teams to interact with their data in group discussions, such as Slack Channels, as well as integrating with many other messaging and collaboration platforms.

The Need for Answers to Ad Hoc Queries

The simplicity of working with a chatbot for your Business Analytics by no means robs you of your Data Analytics power either. Imagine, for instance, that you want to show your management team the latest information on sales figures and the factors driving those figures. Overall sales figures in the blink of an eye, thanks to your chatbot, already help by defining the general context, but you and your colleagues want more. You want to know where the most growth is occurring, which marketing initiatives are having the most positive effect, and where demand is likely to increase, by how much, and whether the business is ready to supply. As your discussion progresses, team members will come up with additional, ad hoc questions as well (if they are all doing their jobs properly!).

Easier and More Powerful Business Analytics

You want to be able to cut to the chase in each case, which is what your chatbot can do for you. Rather than trying to find a predefined view of your data that suits, or constructing each individual calculation, you ask your question directly to your chatbot. Then, instead of being limited in depth and granularity of your analytics, which is often the case for applications focusing only on data visualization, you can – with your chatbot, and the right frontend and backend Analytics System – also immediately explore any detail or facet of your data that is of interest to you.

The Chatbot Choice is Yours

For those who want to continue doing Agile data discovery through dashboards, chatbots are there to enhance, not necessarily to replace. It would be premature to announce “the death of the dashboard”; bots simply offer an alternative that is welcomed by many. Integrated with voice recognition, they can also let you ask a question out loud rather than typing it via a mobile or PC keyboard. The embedding of a flexible bot framework into Business Analytics is a step in the overall journey to simplify complex data and make business insights easily available to business users, anywhere, anytime. As time goes by and chatbots evolve, you will then see that the way you communicate with your Business Analytics technology looks increasingly like the way you communicate naturally with your team and colleagues.

About the author

Ilan Hertz , Head of Performance Marketing at Sisense. Ilan has worked in a variety of management positions in enterprise software companies for the past decade, with a strong emphasis on using data as a means to optimize marketing efforts. At Sisense, Ilan leads the digital marketing team, helping to support the company's hypergrowth through cross-channel, data-driven marketing campaigns. Sisense is an industry-leading business intelligence software company, based in New York. The company provides business intelligence solutions that simplify the process of transforming complex data into actionable business insights. Sisense offers a unique end-to-end solution that lets any business user prepare and analyze data, and then share and visualize the results using built-in dashboard tools.

You might also like...

Is Data Governance Solely About Controls on Data?

Read More →