5 reasons why companies need to develop and prioritize their AI Business strategies now!
Companies need to have an effective AI strategy aligned with business objectives and models to thrive in the digital age.
In 2018, a study by MIT Sloan Management Review exposed that 58% of companies believed AI will significantly change their business models by 2023. And in 2019, a Forbes article that indicated 73% of top U.S. executives have a goal to increase investment in technology dramatically.
Now more than ever, companies recognize that AI is essential to their business growth.
But despite this awareness, which seems to be growing all the time, I still observe that a large number of companies have not progressed beyond the experimental in their use of AI.
Sometimes, when I speak with people in many companies, many of them think of Artificial Intelligence, they often have a vision of something spectacular, almost a realization of science fiction projections. And if it’s not about creating something along those lines, something like an intelligent robot, then it’s as if interest wanes and it loses its meaning.
This romanticized notion ends up precluding the possibility of using something, say, less grandiose, but that provides value to companies and their customers. And the most important — and immediate — value at this moment of technology’s maturity in the marketplace comes from its ability to analyze data and make decisions easier for internal teams and departments and directly for customers.
The current state of the art in AI is what we call “narrow AI,” meaning AI is not intelligent in practice, but it appears to be. This means that one algorithm does one thing well. Combining multiple algorithms with other technologies like RPA (RPA is not AI!) can yield very positive results. So waiting for “intelligent” AI is a waste of time, and the pandemic has shown that time is what we don’t have!
It is all about time…
Today, capturing customers’ tastes, wants, and needs more accurately to enable increasingly personalized offerings is essential to keep businesses competitive.
Consumers have no patience and no time to waste. A person who wants to buy your product has no desire to search through dozens of pages of product offers that have nothing to do with her tastes.
She prefers to enter an e-commerce website or app that offers her a page with a few options tailored to her tastes.
The difference in this context lies in the quality of the filter and its ability to uncover patterns and predict desires. And this is where AI’s ability to harness insights from data comes into play.
Talking about Artificial Intelligence applied to business.
The definition of Artificial Intelligence has evolved dramatically over the years. It can still be described as an attempt to simulate human reasoning ability by enabling computer systems to “learn,” “reason,” “interpret,” and make decisions like humans.
For a long time, Artificial Intelligence was thought to be limited to systems that could solve problems or perform specific tasks, such as playing chess with humans.
But technological advances have shaped the evolution, and today AI solutions are part of a data-driven approach, a culture based on collecting, processing, analyzing, and interpreting data.
These efforts aim to optimize processes, increase competitiveness, and drive better outcomes for businesses and industries.
Today, AI represents a tool that translates data into better experiences, better customer relationships, more loyalty, and therefore a better bottom line for your business.
However, many companies still struggle to see Artificial Intelligence as a valuable tool, a facilitator to create impact rather than impact itself.
That’s why is important to remember that implementing AI in organizations requires more than just hiring data scientists; it requires strategy, organization, evangelism, and education at all functional levels.
It is time to revert the game
We live in a technological moment comparable to the early days of the Internet, which sparked fascination — and even some apocalyptic predictions and fears — when it first appeared. Today, we can all see that the Internet is such a natural part of our everyday lives that we only notice its existence when it’s missing. I’m betting my career that the same will soon happen with AI.
But for companies to benefit from this disruption, they have to stop focus on the technology itself and see it as an amplifier of the value that can be offered to customers, which is at the heart of business strategies.
Many of the discussions about Artificial Intelligence in teams, departments, and companies today still start with the question, “What data do we have, and what technology do we need to work with it?” Please don’t do that anymore!
It’s time to turn the tables and look at what the customers’ needs are. What are the questions your customers are asking that, when answered, will generate value for them and the business? Only when you have this clarity you should start looking for the correct data and related AI applications.
This approach can significantly impact whether or not your AI journey will have successful results.
Let’s imagine that your team is faced with the challenge of improving sales. If your customers’ critical questions are about the best time to buy your products based on specific events in their lives, using AI only with internal company data won’t be enough to deliver the correct answer. With internal data, you’ll likely be able to recommend the best day to buy a particular product, for example (and your customer may not be interested).
It will be necessary to seek external data on their specific events, which, added to internal information about your products and customers’ personal preferences, will provide ammunition for the AI tool to reach its full potential and deliver real value to your customer.
Of course, data scarcity is a fact of life. Not all companies have the infrastructure and money to train sophisticated algorithms with the right amount and quality of data. Therefore, it makes the most sense to start with smaller, less ambitious projects.
In these cases, powerful but simpler models such as linear regression, support vector machines, K-nearest neighbors, and Naive Bayes can be trained with smaller amounts of data.
There are also techniques such as synthetic data generation, federated learning, and self-supervised learning that are very helpful.
Delivering a new vision for AI applied to the business.
We need a new vision that redefines how teams and companies will move out of the experimental stage with technology and use it in practice to generate a decisive impact for the customer and set their business courses.
The pandemic has accelerated things that were already underway. In practice, it brought what was likely to happen in the next five years to the present day.
The pandemic didn’t create new technologies, but the digital acceleration did create a behavior change. I can say that the Covid 19 emergency anticipated many years in a few months. It has made the future come suddenly.
In this context, artificial intelligence came to the fore. We already witnessed many and rapid developments in research and development, but unfortunately, we still witnessed a sluggish adoption and diffusion of AI in enterprises.
A considerable number of companies are still using AI only in experiments in IT labs, developing short terms POCs (proof of concept) and prototypes.
For many of these companies, the “top of mind” AI applications are still chatbots, that mainly focus on automating call centers and service desks.
Of course, a good number of these applications are producing positive results.
I must admit that with the significant increase in demand caused by the pandemic, chatbots have managed to help companies get a faster return on their investment.
However, most chatbots in use in businesses today have an embarrassingly low level for their automated responses. They redirect any more complex interaction than a typical question and answer to human employees, which in many cases ends up in the chatbot not being able to understand the request correctly.
All good, but we can have a lot more from AI.
We need to go deeper.
I firmly believe that AI will only really take off in companies when it will be strongly present on the CEOs, CIOs, CDOs, and the board’s agendas.
AI needs to leave the IT and innovation groups and labs to become an essential topic on the executives’ agenda. But we also need to consider that the technological infrastructure and performance issues cannot be ignored when AI leaves the lab and goes into production. AI demands investments.
It also means that AI governance must be a primary subject of
debate when planning to scale AI across the company.
Due to specific culture and operational characteristics, each organization must define its own model, whether centralized, decentralized, or federated.
There are no cake recipes applicable to all companies equally. Each company will develop its model and adjust it continuously as more and more people gain experience using AI.
Of course, if something positive can be taken from this tremendously challenging moment we are all living during this pandemic is that the digital acceleration it has caused it also started to help to break the barrier of mistrust on AI, and many of us can already see bolder AI initiatives, in addition to the humble chatbots, beginning to appear. In those cases, it means that the issue has already made it onto the board and CEO’s agenda!
The rapid change in user behavior in their shopping habits with the acceleration of e-commerce (think about toilet paper and alcohol gel, at the beginning of the pandemic, followed by food and drinks; later, by toys and, later, by technology and fitness equipment), caused a breakdown in supervised fraud detection, supply chain algorithms and recommendations for customers.
They had not been trained for this scenario! Abrupt changes cause distortions in the responses of the supervised algorithms, as they are not included in your test data.
On the other hand, chatbots, even with their limitations, have managed to help companies deal with the significant increase in demand from users, many of whom are newcomers to digital purchases.
Of course, it is essential to realize that AI is not just a chatbot deployed to improve customer experience. It us a lot more and, it can substantially affect not only the business model but the whole organizations’ operating model.
We are seeing consistent signs that the sense of urgency regarding AI adoption is already being activated.
Gradually, it becomes apparent that AI is not just a technology but also a transformative technology that changes and shapes our society.
And its dissemination is highly dependent on how the executive level considers it a priority and drives its usage as an instrument not anymore restricted to specialists with Ph.D., but accessible in a much more democratic way.
During the last years, I had the opportunity to provide a considerable number of consultancies, to keep valuable conversations and lectures to leaders in several industries, and I have observed many signs of maturity, which are very positive.
In many cases, companies start to worry about data governance, recognizing that AI is not just about algorithms, but data is the primary place to look at.
Leaders are starting to understand that an excellent algorithm, without data, is like an engine without fuel or with low-quality fuel. It will not work.
Also, in many of my conversations (and some of my recent articles), the concept of “Responsible AI,” which involves ethics, concern for data security and privacy, and minimization of biases, has started to be considered in many aspects.
The pandemic has sparked interest in the use of AI. From the chatbots, we will start to see more sophisticated applications permeating the entire organization.
The time to build a Data-driven Culture in business
As we have seen, the implementation of Artificial Intelligence projects depends mainly on the efficient use of the company’s data, generated in large volumes and correctly.
For this to be done, there is a necessity of an update in the organization’s internal culture, putting data at the center of the company’s decision-making and strategic planning.
For this culture to be strengthened within the company, it is necessary to identify innovative and valuable methods for collecting, processing, analyzing, and interpreting large volumes of data. For this, several tools and technologies can assist in this process, emphasizing Big Data and Business Intelligence. The tools in place and the processes must also guarantee the quality of the information and the governance of the data.
As we progressively see that AI is transforming companies’ operations and starting to be part of the “new normal,” it begins to spread and occupy more and more spaces in the company’s operational fundamentals.
The challenge now is not purely technological but cultural. As it accelerates into the core, companies will need to have an effective AI strategy aligned with business objectives and innovative business models to thrive in the digital age.
A highly engaged and innovative AI Strategist. Passionate about communication, with a broad I.T. Management and AI background.