How to crank up the shaft with AI

CRANK up the Shaft

There has been a lot of hype around Artificial Intelligence (AI) and Machine Learning (ML) in the recent years. Whenever there is excessive hype around something, the broader purpose seems to get lost. People tend to use jargon that neither they understand nor do others around them. Hence, it is important to truly understand how to use these technologies to enhance business outcomes and to improve the world. So let us see how we can crank up the shaft using AI and ML.

Why another framework?

The field of AI and ML is indeed highly complex. The engineers deal with large volumes of data and deploy complex algorithms to understand patterns. These details can become overwhelming, even for experts. Under these circumstances, it is easy to lose sight of the big picture.

Why is it that we are using these tools in the first place? Is it because of the Fear Of Missing Out (FOMO)? Or, is it because these tools provide some tangible value?

It is useful to anchor our thinking to a mental model, to be able to think clearly about the purpose of using these tools. AI in itself is very disruptive and can have a wide range of outcomes. Consequently, the traditional frameworks may not apply, when one thinks of it’s benefits. Moreover, the popular narrative around AI is quite lopsided and sensationalist. So, there is a need to have a simple yet powerful framework to CRANK up the shaft of your business or endeavor using AI and ML.

How do we crank up the shaft?

Poor Return on Investment (ROI) continues to be one of the biggest issues with deployment of AI and ML. It is a costly affair and requires significant resources – for obtaining, storing, cleaning and processing data and then for designing, testing and deploying the algorithm itself. If we do not have a clear focus on the value and the end objective, we could easily go down the rabbit hole.

So, here are the five dimensions of benefits that can crank up the shaft – whether in a business context or for the benefit of society. All the components of the crank shaft may not be firing at the same time. However, using the framework can discover hidden opportunities along each of the following five dimensions:

Cost

AI and ML are not the same as automation. However, this is the first image they conjure in the minds of the common man. They are seen as technologies that will help automate, taking away jobs and reducing costs.

While this narrative is only partially true, there is no doubt that AI and ML can facilitate cost reduction. However, there are more use cases where AI can act as a co-worker to help reduce costs, rather than function as an alternate worker.

A lot will change if we start looking at augmenting our processes and workforce using AI, rather than replacing them. Firstly, the attitude towards AI and ML itself will change and this will lead to greater adoption. Secondly, the scope of implementation will increase significantly.

Some people call this artificial intelligence, but the reality is this technology will enhance us. So instead of artificial intelligence, I think we’ll augment our intelligence.

Ginni Rometty (First woman to head IBM)

Once there is a common understanding of how AI can help make our lives better, the economic benefits of enhanced productivity and lower costs will crank up the shaft.

Revenue

Some industries have been deploying AI and ML for enhancing customer experience and revenues. However, in a lot of cases, the focus shifts to reducing the cost of the customer service functions, since it’s measurement is more objective and easier. Consequently, the focus on the top line gets diluted.

Consequently, it is very important to keep a razor sharp focus on increasing the revenues using AI and ML. In many cases, there may be no direct correlation and there may not be obvious metrics to link the AI models with the topline numbers. However, we already know the factors that can help drive growth in revenues for our business. We can use these as surrogate measures for the purpose of building AI and ML models. In a nutshell, we need to become more creative at using AI and ML for increasing the revenues and not just look at customer service and engagement metrics.

How to increase revenues and customer engagement?

But, the golden question that still remains is how do we execute this? So, let me take examples of both kinds of projects that we have deployed, to illustrate with examples.

In one case, we created DiCAP (Digital Content Analyzer and Predictor) to optimize the user engagement with our content. This increased the user engagement that will lead to greater stickiness, customer loyalty and indirectly lead to enhanced revenues. This was also a unique example of using Machine Learning in combination with NLP (Natural Language Processing) to increase customer engagement.

In another instance, we created a ML model to help detect clients whose revenues and other key operating metrics were likely to drop in the coming period. These insights are extremely valuable for any organization, as they can focus on taking preventive measures offline. Typically, the sales and / or CRM teams are victims of cognitive biases. They tend to overlook these factors, even if they are staring us in our faces. This behavior often reminds me of the classic Invisible Gorilla experiment.

Asset

In order to increase ROI (Return on Investment), we can increase returns by reducing costs and increasing revenues. These were the first two elements of cranking up the shaft. The next one deals with optimizing the investment itself. This can be achieved by using the intelligence obtained from AI to increase asset utilization.

There are several examples of enhanced asset utilization across industries, whether it is manufacturing or service. Of course, the concept can be more easily understood with reference to the physical world of manufacturing.

One of the biggest value drivers in any manufacturing set-up is the capacity utilization of machines. Machines have become much ‘smarter’ with IOT (internet of things) devices. This has allowed organizations to leverage AI in combination with real-time feed from the machines. Complex AI algorithms can then be deployed to ensure increased uptime for the machines, increase in the lifecycle of the equipment, etc. These factors are a game changer, especially in industries that are asset intensive.

Asset Utilization in Other Spheres

The possibilities are endless. However, admittedly this is a less intuitive use case of AI, as compared to reducing costs and increasing revenues. One of the reasons is that we are conditioned to visualizing our physical world based on past patterns. Most of the people working in technology are left brained and can solve complex problems. However, they are challenged when asked open ended questions such as “What If”!

New Capabilities

This provides a nice segue into the fourth dimension of the CRANK model to crank up the shaft. We can achieve transformational results if we are able to indulge in divergent thinking and answer these questions that start with “What If”. This is the reason, why I have created a separate section called New Capabilities. It is not a wise idea to constrain our thought process, when we are thinking of possibilities.

Instead of thinking outside the box, creation of value may require us to think as if there is no box.

We can build a much brighter future where humans are relieved of menial work using AI capabilities

Andrew Ng

Let us take an everyday example to illustrate the power of new capabilities.

NLP – What is the good word?

An organization typically has 80% unstructured data and less than 20% structured data. However, most of the traditional focus of analytics has been on quantitative information, i.e. structured data. So, what does the 80% unstructured data consist of and what are the new capabilities are unfolding in this space?

The answer is quite obvious. Isn’t it silly that we never paid more attention to text and language? We communicate in words and not in numbers. However, we compute in numbers and not in words. There are significant advances, mainly in the last decade, that have allowed us to use words as a fundamental unit of computation. This exciting field of Natural Language Processing (NLP) is opening up a lot of new doors that we could not even imagine earlier.

Even today, a lot of people are not thinking of taking advantage of the advances and the possibilities that emerge from this exciting field. And this is despite the fact that we are surrounded by such examples in our daily lives. If you have searched something on Google, you cannot contest this!

The concept of New Capabilities can be extended to other aspects, beyond language. Think of images, videos, audio, music or even the creative arts. Or in the physical space, imagine the new possibilities emerging from the use of robots and drones, and their successors. At a human level, you could think of BMIs, which are Bi-directional brain machine interface that can enable the human brain to interact directly with machines.

Knowledge

This leads us to the final frontier, at least for the time being. There is a lot of speculation about when AI will replicate humans. The opinions are clearly divided on this front. Thankfully, technocrats who work in this field are more pessimistic about the future of AGI (artificial general intelligence), or AI with human-like capabilities.

However, as of now, we are only looking at a framework to help us outline the benefits of AI and ML. In that context, I have saved the best for the last. Today, these technologies can help enhance our knowledge and understanding of the world around us. But, we first need to understand our environmental context before we get to the benefits and value of using AI and ML on the knowledge dimension.

Living In The Information Age

So, it is once again time to put on our creative hats and crank up the shaft. How can we leverage AI to help us become smarter and more knowledgeable? Can we absorb more information in quicker time? How do we filter out irrelevant information? Can we create models that provide us with unique insights? How do we accelerate the process of scientific research?

Nobody phrases it this way, but I think that artificial intelligence is almost a humanities discipline. It’s really an attempt to understand human intelligence and human cognition.

Sebastian Thrun

If you are not familiar with the developments in this space, you may dismiss these as hypothetical questions. However, the future is closer than you think. All of the above questions are being answered by AI and ML in conjunction with NLP. And, to top it all, this is currently a very hot area of research and state of the art is getting re-defined every now and then.

It may be a good idea to view AI and ML as thinking machines, that can help us with our cognitive functions. In fact, the first time the term AI was used at the famous Dartmouth conference in the 1950’s, it was described as a Thinking Machine.

How to crank up the shaft in the future

This historical perspective also helps us look at the future. After the initial euphoria in the 1950’s, AI has had a topsy turvy journey over the years. There have been periods of intense excitement and anticipation, followed by troughs, also known as the Winters of AI. Even today, thought leaders are divided on the future course of AI.

Some of these debates are around AI vs the human brain. Other debates pit the human race itself against AI, warning of potential extermination of the human race. Yet others speak about how AI could replicate human consciousness.

I do not have answers to many of these questions. However, I do have some useful experience of conceptualizing, crafting and deploying AI based solutions that have yielded significant business benefits. I have maintained my focus on using these tools to obtain business benefits. Consequently, some of these projects have been recognized as benchmarks for innovation. Hence, I am distilling my experience in the form of this framework.

Applying the framework to CRANK up the shaft

The trick to using it is very simple. Focus on each of the five elements above and think of AI and ML as just another tool. While I have used all the five dimensions, you may find appeal in only a few of them. Yet, I would urge you to consciously think along each of the five dimensions. Hopefully, you can achieve meaningful returns from your efforts and even exponential returns, if you can exploit new capabilities. It is indeed time to CRANK up the shaft and create the New Normal!

Sign-up for upcoming courses and webinars

In case you are interested in signing up for upcoming courses and webinars to learn how to get business value from AI, ML and Data Science, just fill up the form below and we will keep you updated.

About The Author

Rajiv Maheshwari is a business and start-up advisor, and the co-founder of From The Experts Mouth. He is a management professional with over 25 years of experience, and worked as CEO for a decade, and in leadership roles with NYSE listed companies such as Accenture and WNS.

He is a Chartered Accountant and MBA (Director’s Merit List from IIM Bangalore) and an autodidact, who is on the path of self-directed life long learning and sharing. He is a thought leader, author and keynote speaker and has developed several frameworks to bridge the gap between academia and industry.