For many the idea of an “automated workplace” conjures up an image of a Jetsons cartoon—you know, where spaceships are flying around and robot maids are cleaning and cooking. While advancements in technology has been profound the past several years, the reality is that most workplaces still have hundreds (if not thousands) of processes that are run manually and require heavy-lifting by a human.
In fact, research indicates that nearly 70% of business processes that could be automated are still manual, which was reinforced by our recent study on America’s Most Broken Processes. It’s our belief that the automated workplace will go through three distinct phases over the next two years:
Phase 1: workflow and content automation (today). No-code solutions that empower IT and the line of business to easily automate business processes to increase business velocity, reduce human errors, and drive compliance/accountability
Phase 2: early stage of intelligent process automation (emerging). Workflow and Content Automation (WCA) gets supplemented with process and machine intelligence to not just automate, but also optimize business processes. In these early stages, we believe that algorithms will inform and empower humans to make better, faster decisions. For example, a computer can learn the person best suited to respond to a contract review in the shortest period of time based up on the type of contract, the day of the week, the hour of the day, and even things like the weather.
Phase 3: full intelligent process automation (18 to 24-plus months). As algorithms and machine learning become a standard part of business process automation, they will evolve from augmenting human decisions to making decisions on behalf of the human. For example, you can imagine an expense report approval. Using technology available today, computers can identify patterns that are difficult for humans to sometimes see. For example, if you always approve expense reports for an individual when they are less than $600 and don’t have a single item other than a flight over $300. In this case, the computer could tell the worker what it has learned and ask if the worker wants the computer to always automatically approve expense reports that meet these criteria.
Humans are still good at making decisions and adjusting designs towards more of a human-centric experience. This will be a growing position in an automated workplace.
Preparing Your Workforce in the Age of AI and Automation
There are a few things that are important in terms of preparing your workforce for the future.
First, there is going to be a need to invest in change management to help employees accept that machines will do parts of their jobs. Those employees who embrace this change will flourish, as they will find themselves with more time to create business value by being strategic, creative, and innovative. However, those who struggle to make this change will likely find that they are struggling to find success and satisfaction in their work.
Second, job retraining is going to be important. In some cases, this is being driven by the fact that robots are replacing human tasks. For example, many industries that still rely on legacy computer systems (banking, healthcare, etc.) are using Robotic Process Automation (RPA) to remove or replace people from processes by mimicking their keystrokes. The RPA robot is “trained” to interact with applications via their UIs, using the same credentials a human would—not (or usually not) through APIs and/or web services. In most cases, companies are retraining these employees to perform higher-value job functions that are more directly tied to top-line growth.
What human-centric tasks will AI never be able to replace? I get nervous about using the word “never” when talking about the future of technology and artificial intelligence. Who knows where deep learning will be in a decade or two? However, one thing that is clear is that today, humans need to be active in making the complex, nuanced business decisions that drive a company’s vision, strategy, and culture. For example, machines are great at things like campaign optimization—figuring out how to dynamically optimize a website or a campaign to maximize revenue.
While short-term revenue growth is important for every business, there are often other considerations. You must ask, does making this change diminish the customer experience? Will this optimization help drive revenue in the short-term but degrade profits over the long term? For the foreseeable future (and we’re talking for at least the next 20 to 30-plus years), humans are best situated to make decisions that are strategic, creative, or innovative.
How can companies consider the human experience when developing automation technology? While customer experience has been an important part of technology for a long time, I anticipate it being the major differentiator in the years ahead. Currently, we are living in a world where technology innovation still matters, but that will change over time as more companies get proficient with artificial intelligence/automation. For example, if you compare the adoption and success of Amazon Alexa vs. Apple Home Pod, most pundits view Amazon as having a significant edge given the effectiveness of Alexa vs. Siri in terms of natural language processing and machine recommendations.
However, over time, I see the technology gap closing, which will force companies to differentiate on experience. In the next five years, you can imagine a world where a degree in human computer interface (HCI)will become all the rage, on par with data scientists today. Machine learning and AI are changing every aspect of our lives (from personalized medicine to energy consumption to transportation). We are going to need experts to help us build products to assist with adoption and change management—and ultimately, to optimize human experiences. Already today, I believe that those with a strong HCI background are best-suited to be great product managers.
What is currently being done to push for the human experience among those leading the charge for automation? Recently, at Nintex’s annual conference, Peter Coffee from Salesforce called out the fact that once automation and AI become ingrained into our business and personal lives, we no longer refer to it as AI/automation. It takes on a life/name of its own. For example, we simply “Waze” it to learn the fastest route from Point A to Point B and no longer refer to it as traffic optimization algorithms or machine learning. These naming conventions are good indicators that AI/automation applications have successfully focused on the human experience. If you think about a traffic apps such as Waze, they have worked hard to just make the optimization part of your natural experience when you hop in your car. It behaves just like any GPS system, but it just performs better.
As the workplace becomes more automated, training will become increasingly important for talent acquisition and retention. As workers are freed from tasks that are easy to automate, they will be expected to take on more complex tasks. This will require training, or else a company may have difficulty finding the skill necessary for continued growth.
In addition to embedding AI/automation into our current experiences, other things can be done to improve the human experience. These include overcoming cold-start problems by having users self-identify other people/things like them, enabling machines to leverage the knowledge of others to develop better “out-of-box” experiences until they can learn about your specific preferences (e.g., Spotify makes music recommendations right away, but those improve over time as it learns your unique preferences).
And, of course, as the system learns about your specific preferences, personalization becomes really important. How can I personalize the experience to align with the way that you (or your company) prefers to work? This could include holding non-critical alerts until a certain time of day when you are most likely to act on them, or perhaps recommending a change in workflow to optimize a process, given how things are currently running in your company. Collectively, things like embedding AI into familiar experiences, overcoming cold start challenges, and personalizing the experience to your unique needs will be the key to push the human/machine experience boundaries.