Since the 2007-2008 recession, the drive within businesses has been to reduce operating costs by eliminating labor, and then to replace this labor with business process automation (BPA) that is capable of functioning without human intervention. In other cases, employees continue to hold jobs, but the automation in business processes eliminates much of the routine work, and in some cases, even the decision making process.

Pressed to work “lean,” CXOs and those working for them continue to seek out BPA; they also look to automation to reduce the incidence of human error, and to promote consistency across the organization. Prime examples of BPA include:

  • Fraud detection systems that banks have had in place for years that “learn” the usage patterns of customers so they can automatically issue alerts whenever these usage patterns vary, potentially because of fraud.
  • Fully automated warehouses where robots and mechanized shelves “pull” inventory on demand and then roll it up to a worker who formerly had to do this parts picking himself, walking up to 15 miles on concrete floors each day.
  • Business analytics dashboards that track traffic and fleet usage, and that automatically re-provision delivery trucks to ensure that all service territories are covered.
  • Automated call systems that take users through a series of options and then automatically route calls to the right areas of the business, or that are able to complete transactions with customers themselves — without requiring human intervention.

Systems like these have delivered cost savings and business process benefits to companies, but there can also be a dark side, which begs the question: How much business automation is too much?

As a banking executive, I found this out the hard way when our central system experienced a failure, and we had to failover to manual processes in our branches while we were executing a system recovery. We quickly discovered that there were scarcely enough “old hands” still on board who remembered how to bank with manual ledgers and how to make loans without automated decisioning systems.

The problem is not limited to banking.

  • In retail, if the input to the system of one item code is missed, an employee at the checkout can’t figure out how to manually key in a price.
  • Customer service call systems and phone trees have become so complicated that customers just give up when they can’t get to a real person.

At the end of the day, you begin to run a risk of breeding out so much intelligence from the business with automated processes that you don’t have sufficient trained staff on hand to do the “big jobs” that still require innovative thinking.

We are already starting to see this in IT, with the development of new “instant app creation” toolsets that abstract the system underpinnings of applications because those now doing the app programming work lack the in-depth systems knowledge of their predecessors.

This is not to say that automating business processes is bad — the astounding improvements in productivity that business has made over the past two decades is due largely to BPA. The improvements in cost performance can also be traced to BPA.

Nevertheless, have we really improved the quality of doing business, or the quality of the customer experience for those who do business with the company? These are still open questions.

The danger of going too far with BPA is in its potential to condition employees into mentally lethargic behavior. Then, when a business exception comes around, or a smart analytics report delivers data that fails to synchronize with what is really happening, there is no one there to think through the situation on his own.

This is why the best CXOs are now finding ways to balance the enormous benefits of automation with an equal need to recruit and maintain an intelligent and innovative corps of employees. These CXOs make it a priority to ensure that employees don’t lose sight of the business, and that they understand automated business processes and technologies are only tools.