As workers get overwhelmed with daily tasks, they want virtual digital assistants in the workplace that can alleviate some of the burden.
As life gets busier, knowledge workers are struggling with information overload.
They’re looking for a way out, and that way, experts say, will eventually involve virtual digital assistants (VDAs). Increasingly, workers need to complete myriad tasks, often seemingly simultaneously. And as the pace of business continues to drive ever faster, hands-free, intelligent technology that can speed administrative tasks holds obvious appeal.
VDAs enable workers to use natural language to speak or type commands and queries to complete tasks like searching for information, creating documents or even understanding a data report. Both text- and voice-based VDAs enable more intuitive, even productive, knowledge work. Indeed, according to a 2018 survey of nearly 200 respondents, respondents tracked a range of benefits from AI-enabled capablities: time savings (26%), accuracy (21%) and productivity (19%).
VDAs with a voice interface provide additional benefit by offering hands-free access to information, which is even more appealing to a workforce tethered to desktops, tablets, smartphones and other devices to get the job done.
In the recent report “Virtual digital assistants for enterprise applications,” research firm Tractica estimated that enterprise adoption of VDAs will grow rapidly, from 145.2 million unique active users worldwide in 2017 to more than 1 billion unique active users annually by 2025. Tractica notes that VDAs’ growing sophistication is fueled by significant advances in natural language processing (NLP) and natural language generation (NLG), coupled with equally significant advances in machine learning and deep learning.
But the question is, are digital assistants really sophisticated enough to enhance productivity?
While most investment in bringing VDAs into the enterprise has so far targeted use cases, including customer support, executives are considering how to use digital assistants to boost workforce productivity. But in a nascent market, executives are hard-pressed to find applications of virtual digital assistants that are similar enough to fit their own business processes.
Many enterprises have forged ahead anyway, seduced by the potential power of virtual digital assistants. Early adopters hope to jump ahead of others on a learning curve they know they’ll have to climb one day. Many are also able to solve some immediate problems along the way, reaping some of the productivity gains before their competitors.
And they know their users want voice-based digital assistants in the workplace to alleviate their administrative burdens. Indeed, 62% of respondents to a survey on the workplace expect virtual advisers to become prevalent in the next two years, according to the report “The digital workplace report: Transforming your business.”
“As is the case with anything innovative, enterprises have to balance the prospect of getting ahead against the risks of adopting something new,” said Mark Beccue, Tractica analyst. The risks include the typical risks early adopters of any new technology might expect (cost and schedule overruns, poor user uptake, and vendor solvency). But they also include at least one additional risk that’s specific to AI: Vendor lock-in is particularly dangerous with any system that learns from your users, because replacing one system with another means starting the learning process all over.
To maximize the advantages and minimize the risks of being an early adopter of VDAs, the most important step is to carefully choose your first use case. The ideal initial use case is one that’s easy to implement and that promises an easy return on investment.
For the time being, it’s safe to assume that virtual digital assistants in the workplace can handle only question-response interactions, as is the case with virtual digital assistants such as Alexa, and Google Assistant. You can ask a VDA to give you the phone number of John Smith or tell you the status of your largest sales opportunity. But you can’t tell a VDA to book you a ticket to New York and expect it to do all the tasks involved in making travel plans, such as checking availability, comparing flights, and so on.
So far, scenarios in which digital assistants in the workplace enhance productivity fall into three categories: scheduling, project management, and improved interfaces to enterprise applications. “Using digital assistants to perform scheduling has clear benefits,” Beccue said.
“Scheduling meetings and managing calendars takes a long time—many early adopters are able to quantify the savings they get when the scheduling is performed by a VDA. Likewise, when VDAs are used to track project status through daily standup meetings, project managers can easily measure the time saved.”
While the third category for voice-based digital assistants in the workplace isn’t widespread yet, using VDAs to improve the front ends of enterprise applications will likely become the dominant use case in the not-so-distant future. By 2020 major business application players will begin to dominate the business application VDA market with their own voice-based assistants as front ends to their own business applications, according to Tractica.
Major business application vendors are already onboard with artificial intelligence and digital assistants as front ends to AI-enhanced business applications. Admittedly, though, products are immature. After all, while intelligent CRM has become a buzzword, the current generation is still light on intelligence.
As you choose a use case, develop a vision about how you might use digital assistants in the future. Future generations of NLP front ends will be able to contextualize, similarly to how humans rely on context to understand meaning. For example, if you work for a big company and you ask a digital assistant to list your competitors, the VDA has to guess that you are asking about competitors to the products you sell in your department, as opposed to competitors to other product lines in other parts of your company.
Another area where VDAs are sure to improve is in the use of machine learning-based predictive analytics to spot patterns and make predictions. Not only can VDAs help with predictions, but also they will become proactive and warn users without being asked to do so.
But future VDAs will be useful only if they have large volumes of high-quality data to learn from. The major vendors that offer VDA as a service already have huge pools of data from which their digital assistants learn. Their digital assistants have already become adept at natural language processing through the vast number of interactions that are only possible with a very large user base.
Large enterprises may be tempted to develop their own digital assistants, because they don’t want to trust a vendor with their data. If enterprises have the resources to develop a solid current-generation virtual digital assistant and they have a big enough user base to provide learning data to the digital assistant. The downside to this path is that, to continue to be useful, your custom-made VDA has to evolve with other VDAs in the market. How will custom software keep up with VDAs that complete even moderately complex tasks, such as “Book me a ticket to New York” without maintaining world-class expertise in-house?
Perhaps the most important change we’ll see in future generations of VDA technology for workforce productivity will be the advent of general-purpose VDAs that help users with all tasks. These VDAs will be multi-channel (providing interfaces through mobile apps, messaging, telephone, and so on) and they will be bi-modal (enlisting text and voice).
The general-purpose virtual digital assistant hasn’t arrived yet, but the technology is heading in this direction. Early adopters of the current-generation of VDAs are more likely to be prepared for the future generations, especially if they start with the right use case.
Affiliated professor at Grenoble École de Management, and author of the book Master the Moment: Fifty CEOs Teach You the Secrets of Time Management, Pat Brans writes and teaches about cutting-edge technology and the business surrounding technological innovation. Previously, Brans worked in high tech for 22 years, holding senior positions in three large organizations (Computer Sciences Corp., then-HP, and Sybase).