AI And Digital Workstyle Analytics Can Provide New Insights
When we think about the analysis of our workplaces, we are typically led toward the ubiquitous “employee survey.” While such surveys are not without merit, there have inherent shortcomings.
First, getting employees to respond is always a challenge. Often, those we most want to hear from are those least likely to respond. Second, how people think they work and behave may not reflect reality. Third, because of the effort required to conduct them, employee surveys tend to be broad-ranging and high-level. This is reflected in the insights derived and, therefore, the improvement initiative proposed by them—broad, large and expensive.
The good news is that increased digitization and a global pandemic that forced a majority of staff members to work remotely have accelerated our ability to identify work patterns and behaviors for all employees. Regularly called analyzing the “digital exhaust” from staff members’ online activities, we no longer have to rely on survey sampling.
We can analyze how all staff members are working and collaborating online (or not). We can identify how formal organizational units are interacting or not. We can identify those individuals who are acting as “collaboration champions,” either connecting across disparate groups (bridges) or bringing staff together to work as a cohesive whole (bonders). Aided by AI, workstyle analytics has the potential to provide brand-new insights into how we work.
Gartner Inc. identified workplace analytics as one of the emerging technologies from 2021, but it’s not without its challenges. I have previously addressed the issue of privacy and ethics. Gartner indicates that employees are open to having their work patterns monitored if there is something in it for them (e.g., identification of training sources that may help their career advancement, provision of useful information).
Microsoft AI Report Exposes Survey/Workstyle Analytics Contrasts
A May 2023 work trends report from Microsoft focuses on the potential for AI to relieve staff of the drudgery of routine “busy” work, leaving more time for creative and innovative work that leverages their innate “humanness.” AI, of course, needs workplace data—in this case, LinkedIn Economic Graph data and content. The study drew its insights from staff surveys and Microsoft 365 productivity signals (workstyle analytics).
The survey results reported that “the number one productivity disruptor is inefficient meetings, followed closely by having too many meetings at number three.” The workstyle analytics results found that “the heaviest meeting users (top 25%) spend 7.5 hours a week in meetings”—or 1.5 hours a day. Our own workstyle analytics results show 1.6 hours a day for the top 25%.
Is 1.5 hours of meetings a day for the busiest 25% of staff too much? We looked at the percentage of staff that was involved in more than five hours of meetings per day (including one-on-one calls) that one might consider real overload and found that the number is less than 1%.
Even a single inefficient meeting a day might be seen by some as one too many, but are we overlooking something? What about the “long tail” of under-collaborating staff? They are the “quiet majority” who likely don’t respond to staff surveys. Patterns of digital interactions have followed this pattern since the early days of the internet (recall the 90/9/1 rule). This pattern is sustained to different degrees in all of our digital workstyle analytics work.
The economics of digitization allowed Amazon to successfully sell “long tail” books to the masses. Could it also help provide support to the “quiet majority”?
Using AI To Address The “Long Tail” Of Non-Collaborating Staff
Imagine what an impact AI could have in bringing the long tail of non-collaborating or low-collaborating staff even a few percentage points closer to the highest collaborating 2% to 3%? Perhaps a nudge from the AI co-pilot to poorly interacting staff members in groups or teams with higher collaborating members. Perhaps a subtle reference to high-collaborating individuals in similar roles. Maybe some help for overworked leaders, identifying precisely those groups or teams that are overly siloed or not working cohesively and in need of their attention.
Practical Steps To Take
There are several steps you can take to start to understand the power of the workplace data you already have even before investing in sophisticated AI-based products:
• Your IT administrators have access to the raw administration reports on workplace use (e.g., time in meetings, email activity, enterprise social activity, chat etc.). Ask for permission to peruse these reports to identify the base signals that may be useful to the business.
• Knowing the basic signals that are available, think about potential insights that would be useful to the business. For example, what proportion of staff are least digitally active? Which departments meet online the most?
• Think about where online discussions are happening in your organization. Are they happening in Teams channels? Enterprise social networks? Chat? Email? Which sources do you think best reflect the culture of the organization?
• Finally, if your company directory is in poor shape, start remedying this now. The more accurate and up to date your staff profiles are (e.g., departmental membership, work locations, gender, roles, etc.), the more effective the AI and workplace analytics solutions will be.
Conclusion
AI-supported workstyle analytics promises to democratize investments in workplace improvements. The quiet majority of low-collaborating staff will receive equal attention to the minority of overloaded staff. The collective result? An equalization of workloads, less friction, more flow, less stress. Who wouldn’t want that?
Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
Read the full article here
