I have been closely following the industry evolution, and discussions on artificial intelligence (A.I.) for a while now. One theme keeps on coming over and over, BOTS!
What are bots?
Answer: We are trying to replace humans by 'machines' aka bots that would talk to us like a human (hopefully!) and deliver automation. Machines are the various industry frameworks, platforms and engines powered by Google, Facebook, Microsoft, IBM (Watson), and so on. A machine mimics "cognitive" functions that humans associate with other human minds, such as "learning" and "problem-solving."
Bots phenomenon isn't new, and it did exist for over a decade in the form of virtual assistants to assist customers. The key challenge with virtual assistants had been the lack of intelligence. A virtual assistant could only scan through a predefined knowledge base and provide a response to the customer, mostly on the website or through an app. Most of the times the answers were not helpful or out of context.
The close study of virtual assistants gave rise to the rapid advancements in the Natural Language Understanding (NLU) and Natural Language Processing (NLP). Bots framework delivers intelligent business services through:
Intent detection: This is the most critical element of any conversation i.e. what is asked, and how should a machine respond? The NLU/NLP gives machines the capability to detect 'intent' in the question posed. It does require initial modeling by humans to build, test, learn and optimize NLU. Thinking: Once paired with A.I., machines can 'think' to provide the best possible outcome. It also starts with training machines with possible outcomes or answers based on the 'intent.' The process does require humans to train the machines and help them build initial confidence. Deep Learning: Once machines can understand, learn and deliver the outcomes, A.I. further powers machines to continually learn, deeply analyze and connect customer information to build context for various conversations and personalization.
MICHEL TCHEREVKOFF VIA GETTY IMAGES
Welcome to the future!
Are bots going to replace humans? My single answer is 'NO.' Although we would love to believe it, we aren't there yet. I don't see this happening at least in the next three to five years.
There is a deep connection between bots and the humans. The ideal customer experience offering cannot just live with a bot. People are complex and carry emotions, and customers do require that connection. Once a customer reaches a point of frustration, we must hand over the customers to a human for help. Coming few years would see multiple use cases of (a) bots assisting customers and (b) bots assisting humans to help customers.
But why bots will become increasingly important and replace apps?
We use only 3-5 apps on a regular basis everyday. Various industry point of views state that apps have failed us and reasons are quite obvious:
Firstly, an app requires learning, and it has a user interface that becomes complicated with each new release.
Secondly, an app discovery is trivial. One needs to know the name of the app, or it should show up as a 'top-ranked.'
Lastly, an app has too many friction points (sign-up, sign-in, and remember to use it again).
All this cognitive load has put us in the 'app fatigue.'
This is the opportunity for bots:
The bot user interface is super easy; it's a no brainer. Everyone has used some chat or text interface
There isn't any discovery; a bot can live anywhere from website to social media to your messaging app
No friction points i.e. there isn't any download, sign-up or sign-in required. The bot is available to help where you are
Once the bot can learn about you through questions, purchase history, or social media accounts, it can personalize content.
The challenge we have with bots is to go beyond 'one off' use cases and get into the 'contextual conversations' to deliver the right level of personalization for the customers.
Venture Beat (Stefan Kojouharov) predicts the bots to dominate in the following areas:
Retail: Imagine being able to quickly find out that Sephora just got the Selena collection in stock, and then be able to put the signature red lipstick on hold. Bots will offer this level of personalized experience without the heaviness of an app.
Customer Service: Everyone hates calling 1–800 numbers, and soon bots will become a much-loved alternative. Imagine never having to be on hold again!
Local Services: Bots will be great at getting local services. Need to preorder lunch, book reservations at a hotel, or even hire a local plumber? Bots will automate this process.
Scalable Personalized Content: Bot can segment their audience and give different users different experiences. This is something you can now do with Chatfuel, and it’s showing signs of great success.
Social: Soon you and your friends will be able to summon a bot to group chat to have it order movie tickets, book a trip, or even order takeout.
One-Time-Use Cases: In areas where the epistemology is well known, a bot will be able to offer professional-grade services. For example, a dermatology bot will be able to monitor your skin ailments and prescribe medicine. Bots will be able to quickly scan your eye and tell you if you have diabetes (something Google has done successfully) or even fight a parking ticket for you.
Conversational: When multiple variable inputs are needed, bots excel. This means that bot apps can more effectively understand a particular need than an app can. Furthermore, conversations are very human, which means that a bot has a much higher chance of being memorable.
We all know that A.I. and bots have a long way to go and for some companies, developing a bot is not a high priority. But for those with high-volume customer service centers, those wanting to increase sales productivity, automate marketing tasks, and for simple scheduling tasks, bots make a lot of sense.
Application of bots and A.I. goes beyond just text - into voice, image & facial recognition. We are only scratching the surface!