menu

Intelligent Empowerment: The Next Wave of Technology-led Disruption

Technology is leading a new wave of disruption in our society. While powerful governments are worried about the potential implications of “intelligent” systems and robots displacing jobs, we're seeing more examples of such systems enabling business transformations. Intelligent Empowerment is a shift that brings together the best of both worlds: augmenting human intelligence with machine intelligence through the use of data and techniques such as Optimization, Artificial Intelligence, and Machine Learning.

The market opportunity is real, with reports of AI being a 15 billion dollar industry, projected to rise to over 70 billion by 2020. IBM alone is investing 3 billion dollars to bring their Cognitive Computing to the Internet of Things (IoT). Samsung is acquiring Viv, the next generation platform powering intelligent assistants. Google is hiring top talent from the Artificial Intelligence field and redefining itself as a Machine Learning-first company. Big players are partnering to formulate best practices and to open source many of the frameworks, algorithms, and tools such as TensorFlow and NuanceMix used to implement AI and Machine Learning solutions. These are just a few of the signs that the shift is already happening.

This begs the question: how can you best leverage Intelligent Empowerment to transform your business? 

The promise

A lot of these techniques are not new. The AI field has been around since the early days of Computer Science, but why is it gaining so much attention again now? There's a combination of factors that contribute to this renaissance: more access to data, increased computing power to store and process the data, advances in the algorithms and techniques, and the increase of Open Source tools that help lower the barrier to adoption. Combined, they enable the creation of innovative solutions and products.

We can look at the automotive industry for an example of this revolution: cars have had sensors for a long time, but they have only been used to provide real-time data about the car’s own health—for example, current speed, fuel levels, temperature and tire pressure. With the growth of IoT, a connected car could now be part of a wider ecosystem, combining its own data with other datasets to enable intelligent navigation, detection of traffic patterns and accidents, and allowing manufacturers to analyze usage to encourage predictive maintenance. The current trend is now the intelligent car that combines the benefits of sensors, connectivity and advances in AI algorithms to enable self-driving cars.

If we apply the same lens to another domain, such as supply chain, we can see a similar progression. Initially, supply chain solutions focused on improvements at the enterprise level. Now we see a trend towards integration and a more responsive supply chain that can participate in a wider connected ecosystem. In the future, the intelligent supply chain will be proactive in self-healing when disruptive events occur.

a chart showing technology trends from past, present and future. Past trend: Isolated, Self-centered, Reactive. Current trends: Connected, Collaborative Ecosystem, Responsive. Future trends: Ubiquitous, Intelligent, Proactive

Intelligent systems promise to bring great benefits to your business such as productivity and efficiency gains, enabling growth and scalability, and reducing costs to your normal business operations. They can also have a wider impact on society such as better detection and treatment for diseases, helping humans live better and longer. On the other hand, it can transform the workforce and automate away many people’s jobs.
"In a post-digital age, Intelligent Empowerment offers a new lens to look into the future."
Tweet this
The digital revolution has enabled the increased connectivity powering the current trends, but in a post-digital age, Intelligent Empowerment offers a new lens to look into the future. However, in order to avoid making the same mistakes, it’s always good to look at history to understand how previous technology innovations made their way into the enterprise.

A brief history of technology innovation

We have seen many waves of technology-led business disruption. Computing first made its way into the enterprise scene in the form of mainframes that would automate basic computational functions, such as accounting. With the advent of the PC, we have seen a shift towards decentralized computing and a shift of focus from hardware to software. This second wave brought along more business process automation and saw the rise of enterprise systems such as ERPs and CRMs.

technology-timeline-by-danilo-sato

The third wave of disruption was led by the internet. System limitations such as the Y2K bug and the introduction of a new currency (the euro) accelerated the replacement of legacy mainframes and systems with web-based applications that offered a richer user experience. New business models were now possible, and we have seen the growth of companies such as Amazon, Google, and Facebook. This also enabled the creation of large data sets, with more and more users connected to the web.

The launch of the iPhone drove the next disruptive wave through mobile. It brought widespread accessibility to the internet and a new channel to consumers, along with a new distribution model through the app stores. Suddenly companies had to invest in building their mobile apps to reach their customers where they were.

The proliferation of digital devices with increased connectivity—the Internet of Things—and globalization of the workforce drove the fifth wave of disruption: digital. With more channels to reach consumers and their increased expectation on how to interact with organizations, every company invested in efforts to become digital enterprises. Beyond the external impact, digital also had an internal impact on employee behavior, and digital transformation became high priority on the leadership agenda.

While the initial waves were about simple automation of existing business processes with technology playing a supporting role to the organization, the recent disruptions are bringing technology closer and closer to the core of every business. As one revolution ends, the powering technology doesn’t go away, it just becomes so widely adopted that it can be taken for granted.

Intelligent Empowerment is the next revolution after digital. Every company wants to become an algorithmic enterprise and bring intelligent systems into their organizations. However, in the same way this happened with mobile and digital, this won’t be an easy transition to realize the full benefit. It goes beyond technology and requires a transformation in people and processes of equal magnitude.

The reality

While the introduction of mainframes took 20-30 years to take place and companies had another 10 years to fully adopt PCs and replace legacy systems, the time to adopt new technology is shrinking. The internet revolution rose in the mid-90s, and even after the dot-com bubble burst in the early 2000s, web 2.0 pushed companies to adopt web-based technologies in a shorter timeframe. Mobile rose in 2007 with the launch of the first iPhone. By 2011, John Paton had coined the term “digital first”. This means an organization’s time to adopt and incorporate new technology is shrinking.

With the intelligence revolution, we are noticing a pattern with our enterprise customers. In order to cope with the speed of innovation, companies are looking for quick wins in the form of point solutions and off-the-shelf products that tackle intelligence in a specific area. They improve productivity or efficiency in small pockets. The same approach was taken with previous innovations such as the internet and mobile, and we’ve seen our customers immediately create legacy websites and mobile apps before realizing they needed a more cohesive digital strategy.

Here are a few examples. A retailer is looking at smart algorithms to improve various supply-chain and planning functions: an inventory planning algorithm, a logistics and transportation algorithm, a sales forecast algorithm. Another leading financial services client is looking at solutions for fraud detection, know-your-customer (KYC), and transaction surveillance. There are products and companies in the market claiming to solve such problems using machine learning and AI.

The problem with this approach is that it's sub-optimal, incurring significant integration costs. As any Enterprise Architect could attest, bringing new products and technology solutions into an enterprise is not an easy task. They are usually additive and contribute to increased complexity in the systems landscape. And when they have to integrate with existing systems, being able to fully test and deploy them into the enterprise ecosystem could take years.

We believe a holistic approach is a better solution. Instead of point solutions that optimize locally, consider how intelligent systems can help optimize your entire organization as a whole. The bigger problem is how technology can improve the speed of the overall business, and big efficiency and productivity gains will only come when you break out of existing systems and organizational silos.

We have been working with a large customer in the supply-chain industry trying to apply holistic intelligence to their planning capabilities. A truly end-to-end planning system goes beyond the existing silos of production planning, material planning, demand planning, transportation planning and so on. It requires an approach that links those separate domains to drive a global optimal solution. It also requires a data platform that supports a responsive streaming architecture, enabling the system to go beyond daily/weekly batch processes into one that can react at the speed of business.

From a systems perspective, we are not simply looking at adding a new product into their landscape, but how to replace the existing planning modules, COTS products, and add-on's that were built around their existing ERP system. From a people perspective, it’s redefining the role of the planner, since the new system can provide an overall view of the entire supply-chain, rather than specialized roles developed to manage and operate the existing silos. From a process perspective, the new system can react and respond to disruptive events during the course of the day. Rather than being fully driven by humans, the system needs to raise incidents and propose solutions that the human can evaluate and execute. Over time, our system can learn from data and make routine decisions on behalf of the user.

We believe that to really drive Intelligent Empowerment into your organization, you should think beyond the quick wins and consider a more holistic approach. Although it will require a change program to transform people, processes, and tools, the potential benefits far outweigh the costs.

How to get started

Based on our experience discussing and bringing Intelligent Empowerment to our customers, we came up with a few questions to help you evaluate if your organization is prepared for the next disruptive wave.

Does your data strategy support Intelligent Empowerment?
 
Data is at the core of these new intelligent systems. The infrastructure has to support storing large volumes of data and a scalable processing layer for different types of workload, from batch to more real-time and streaming needs. A modern approach comprises of a streamindataflow architecture, with microservices that can react and produce data streams that flow through a distributed and scalable event bus, using something like Kafka as the backbone. It also includes an Enterprise Data Lake that can be used for long-term storage of raw data from multiple sources that can be further used for Data Science and enterprise analytics needs.

We have seen many clients investing on Big Data infrastructure, without changing the organization structure and building the skill-set required to fully leverage new Data Science and analytics applications. That’s why SQL on Hadoop has become a popular topic. However, to fully leverage Intelligent Empowerment, you will need to go beyond batch-oriented thinking and traditional data warehousing techniques and bring data thinking into the core of your system architecture.

But the technology foundation is just the starting point for riding this disruptive wave. There are other questions you should think about.

Are you ready to break away from existing silos and think holistically?
 

To really become an algorithmic enterprise, you will need to reconsider the organizational structure and re-evaluate if the existing silos are really needed to take your company to the next level. Applying intelligent solutions to a small problem might be enough to get you started, but you should really take a holistic approach that spans across the existing organizational structures to realize the full potential.

This requires rethinking not only how people are organized, but also your existing processes. With intelligent systems being able to provide data-driven insights and perform data gathering and processing tasks much more efficiently than humans, you need to consider what roles will require real human power in the future. There will still be a need for human intuition and creativity in the workforce, but the predictable and procedural activities of some jobs will be more easily replaced by automation.

As exemplified by the supply-chain story, they had to create a new role that can oversee the entire value stream. This is a higher-level position that will likely require new skills, such as understanding how the Data Science and optimization algorithms can be tuned, but also shifting their current focus. Instead of thinking about the present and trying to solve disruptive events, the system will help them identify those issues and propose solutions so they can spend more time analyzing past data and performing future-looking simulations; something they can’t do today because of the current workload imbalance.

What business problem are you trying to solve?
 

Finally, we strongly advise you to consider the business problem you are trying to solve with Intelligent Empowerment. It’s important to go beyond just riding the hype cycle. You don’t want to end up in a situation where you are looking for problems that the intelligent solution you purchased could solve.

When done properly, Intelligent Empowerment can help lower your total cost of ownership from existing systems and products, as well as deliver productivity and efficiency gains on how you operate your business. Bring your Enterprise Architects along in this journey and make sure they understand the business problem you are trying to solve. It will not only help simplify your systems landscape and retire many legacy systems, but will also help modernize your overall architecture. Since this will take time and require changing technology and people’s mindsets along the way, we believe taking an evolutionary approach to architecture can help offset the risks of trying to implement this in a big bang approach.

Intelligent Empowerment has the potential to disrupt your organization and modernize your architecture, though we recommend thinking beyond the technology enablers. We can learn from our history of technology-led disruption and avoid making the same mistakes when bringing them to the enterprise. Take a holistic approach and consider the overall technical, people and process implications to really leverage its potential to transform your business.