With the five recommendations of their AHEAD model, the authors (Cognizant’s Malcolm Frank, Paul Roehrig and Benjamin Pring in their new book: What to do When Machines Do Everything) lay down a path for companies to harness the power of machines and gain competitive advantage, or just plain survive.
The first step in the corporate journey is to automate as many processes as possible. Managers are encouraged to think big: cost reductions of at least 25 percent and similar productivity gains. Only such radical thinking will enable a company to break the inherent inertia of naysayers and other opponents to change. Recommended areas for automation are repetitive back or mid-office tasks, namely in areas such as finance, accounting, IT, customer service, human resources, etc.
Trizetto, a US company specializing in healthcare administrative processing, is an example of successful automation, providing assistance to insurance companies and healthcare facilities in reducing “overhead” costs. So successful that the company was acquired by Cognizant (the authors’ employer) for US$ 2.7 billion in 2015.
Yet without data, no automation is possible. So the authors’ next recommendation is to create a “code halo”, that exhaust trail of data that enables machines and software to learn and to crunch. Sectors such as financial services or healthcare generate data in industrial quantities, but even other businesses are learning fast. The key here is often to place physical sensors in equipment, so as to monitor performance.
General Electric manufactures locomotives (e.g. the Tier 4 model) that are true rolling data centers, with dozens of sensors installed in key components. Minute improvements in operations (“..a bit slower on that throttle, Scotty..”) or in full fleet coordination can mean millions in savings to rail operators.
Step 3 in the machine tango is to enhance the customer experience by incorporating machine-based improvements. A simple case in point are modern automobile GPS systems, which not only provide maps, but also are increasingly able to steer drivers clear of traffic jams, accidents or roads under repair. Enhancing the customer experience by using accumulated data is what Amazon or Netflix excel at, when suggesting books or movies to known customers.
The authors turn to education for an example. McGraw-Hill Education, via its ALEKS knowledge system, is redefining how teachers can work with large classrooms. Since ALEKS can rapidly assess student’s levels, strengths and weaknesses, it can prove very valuable to teachers, who can then focus on targeted remedial action, and spend less time on routine tasks such as grading or collective teaching.
Investing in machine capacity – whether for software, sensors, data analysis or other uses – makes no sense unless there is some return on investment. For the authors, this return can be in terms of sometimes drastic cost reductions, and therefore price reductions to customers. That is where abundance kicks in. Henry Ford’s assembly lines chopped auto prices several-fold, making them accessible to the general public. Textile looms made clothing affordable. In similar fashion, FinTech, HealthTech or EduTech can generate similar revolutions in their sectors, meaning threats to established companies. The clear message from the authors: watch your back or your company could be toast when an upstart rival takes over.
Needless to say, innovation (“discovery”) is another important stage along the road to machine-compatible business. The authors remain largely skeptical about direct uses for software, artificial intelligence or other bots in generating corporate innovation, but they do emphasize that this is one area where human capabilities are critical. Yet machines can prove useful in the “kaizen” innovation mode: frequent, incremental innovations that come both from machine lessons and from human intuitions.
Where does this leave us?
AHEAD provides a simple framework for understanding how machines (software, AI, bots) can help companies reduce costs, increase productivity and simply survive. Most importantly, in the eyes of the trio of authors, machines might terminate some jobs, but for the better. In their stead will pop up new opportunities for more qualified skills. As they point out, who in their right mind would now want to be a toll collector on a highway or a garage attendant in the middle of the night? Data miners are far better off!