Why Remote Doesn't Work
A framework for predicting the effectiveness of remote work
by Greg Lemiska and Kevin McManus
For millennia, humans have gathered to collaborate on all manner of tasks and typically benefited from doing so. It is the essence of the human experience to come together face-to-face, work in unison, exchange ideas, and share the fruits of our labor. Modern people have worked together in factories and offices to collaborate on previously unimaginable problems as they discovered, designed, and built engineering marvels, innovative products, and complex systems.
During the COVID shutdown, essential workers were forced to continue working in person despite the risks, while knowledge workers needed to find ways to make a living without personal interaction. Remote work suddenly became ubiquitous for knowledge workers in lockdown.
Organizations were quick to proclaim the benefits of remote knowledge work, including increased flexibility, reduced costs, and enhanced productivity. For some, remote work is a fundamental necessity to the business model. However, as time passed, other organizations saw burnout from longer hours, quiet quitting from lack of engagement, and ultimately, drops in productivity.
Fast forward years after the end of the public health crisis, and most companies are still trying to calibrate their remote work policies for the best results.
Consensus on remote work for knowledge workers is changing
There is an emerging nexus of US CEOs sounding the alarm that productivity is declining in a remote environment. Salesforce CEO Mark Benioff recently messaged his organization via Slack, asking for suggestions on the root cause of productivity loss.
“How do we increase the productivity of our employees at Salesforce? New employees (hired during the pandemic in 2021 & 2022) are especially facing much lower productivity...” (1)
A study by economists at the Federal Reserve Bank of New York, the University of Iowa, and Harvard University found that while remote work may boost output in the short term, one remote worker can reduce collaboration between the team as a whole. (2)
Mark Zuckerburg, a once vocal proponent of remote work, has changed his tune significantly. Recently he talked about “performance data” that showed in-person engineers performed better than those who were remote, and Meta is backtracking on some aggressive remote work policies. (3) Returning Disney CEO Bob Iger put it like this:
“In a creative business like ours, nothing can replace the ability to connect, observe, and create with peers that comes from being physically together, nor the opportunity to grow professionally by learning from leaders and mentors.” (4)
Even workers themselves report dissatisfaction, with 53% saying it is harder to connect and interact with coworkers (5). More broadly, a recent New York Times article highlights the fact that results are highly variable:
“Some papers have linked remote work with productivity declines of between 8 and 19 percent, while others find drops of 4 percent for individual workers; still other research has found productivity gains of 13 percent or even 24 percent.”
Remote work was a boon to organizations and workers during the pandemic, but it is clear that data and opinions are shifting toward the idea that in the post-pandemic world, not all scenarios are a good fit. How does an organization determine the correct approach?
Interpersonal interactions drive the difference between remote and in-person productivity
At the core of the remote problem is the difference in interactions between people in a remote vs. in-person environment. Specific behaviors benefit from person-to-person interactions and are less effective in remote scenarios. The table below outlines the types of behaviors that are challenging in a remote context.
This list will make logical sense to managers, some of whom will manage teams or individuals who are good at performing them remotely and some who are not. In many ways, the essential behaviors of management & leadership are undermined by a remote environment. In order to predict remote results, It’s useful to understand how often these skills/behaviors need to be employed for a particular role.
Three key factors determine the frequency and importance of interpersonal interactions
If interpersonal interactions are the primary drag on the efficacy of remote work, the key factors that determine if a role or task is suitable for remote work drive how often interpersonal interactions are required. These three factors are the “wickedness” of the work, the degree of process maturity achieved, and the engagement level of employees.
1) “Wicked” environments require more interpersonal interaction.
“Wickedness” was first used in the modern sense by C. West Churchman in 1967 (6). David Epstein describes a more simple and broadly applicable definition in his 2019 book Range. Epstein describes the terms kind vs wicked as follows:
Kind environments are those with clear rules and feedback, where repetition leads to improvement through accurate and immediate feedback.
Clear and consistent rules
Wicked environments are complex, ill-defined, and have no clear solutions. They are often characterized by the following challenges:
Unclear or incomplete rules
No obvious or repetitive patterns
Delayed or inaccurate feedback
These wicked challenges are often overcome via team interactions, knowledge sharing, coaching, collaborative brainstorming, distributed experimentation, etc., which suffer from the remote challenges outlined above. There are certainly cases where wicked problems can be solved by an individual. However, it is more common that wicked problems drive the need for interaction with other people who have additional, adjacent, or even seemingly unrelated knowledge. Therefore, the more wicked an environment is, the more often the behaviors outlined in Table 1 are needed, and the more negative impact remote work will have on productivity.
It then follows that the frequency of wicked problems should influence decisions about remote work policy. Below, we outline a five-tier scale for evaluating the wickedness of a given role:
Wickedness is a catchy term but not an entirely new concept; business has created tools and techniques to reduce wickedness in various business processes for decades. Lean focuses on waste, and Six Sigma focuses on variation. But whatever we call it, business organizations are always attempting to reduce wickedness. These efforts lead to more mature business processes, policies, and procedures. Successful organizations clarify rules, increase repeatability, monitor KPIs, create feedback loops, and eventually create predictable outcomes. Thus, process maturity is our second factor driving suitability for remote work because of its direct impact on the frequency of wickedness.
2) Low Process Maturity increases wickedness.
Workers everywhere have experienced otherwise simple processes becoming frustratingly complex due to vague responsibilities, unclear performance targets, or undocumented process steps. Tools, training, and analytics used to support and manage processes reduce the wickedness experienced during the execution of work. A peak performing process might be said to have achieved “minimum wickedness” for that work. One framework for measuring how successful an organization has been at supporting a process to minimize wickedness is ISACA’s Capability Maturity Model Integration (CMMI), which includes five levels of increasing maturity.
Employees working with processes on the low end of the scale will encounter wickedness more frequently than with processes at the high end. In the absence of formal efforts to increase maturity, organizations force individuals to create “tribal knowledge”. Most workers will have experienced a situation where one individual is the “go-to” person for some process because of their painfully earned experience. New workers attempting the process experience wicked problems because they lack tribal knowledge, and the team interactions needed to get through it require more effort in a remote work environment.
This highlights how process maturity is particularly important to new or inexperienced workers. How much of a job or role is tribal knowledge vs. part of formal onboarding and training? How clear are performance expectations for a new worker who does or doesn’t have a KPI dashboard for their role? How much wickedness will they encounter in the first weeks, months, or years in their roles?
This brings us to the final factor impacting the efficacy of remote work. As employees seek to adapt to new roles or just perform old roles, how do these challenges impact morale & motivation?
3) Low engagement increases wickedness.
A disengaged worker works fewer hours, is less effective during those hours, and is more likely to give up when things get difficult. In the worst-case scenario, disengaged workers engage in behavior like “quiet quitting”. Working on a team with a disengaged worker (or perhaps even more than one) creates extra friction not just from the work that that person does not complete but also for every other member of the team who needs to engage with that person. Disengagement creates even more difficulty in interpersonal interactions in a remote environment.
The impact on productivity from individual mindset, motivation, temperament, coaching, and other elements of the work environment may be the single greatest factor determining the performance of an individual remote worker. It’s worth noting that the factors and level of effort to engage an employee effectively will be highly individual, and some folks may thrive. However, a large portion of the workforce will be negatively impacted by the challenges to interpersonal interactions we’ve discussed, and wicked domains are more likely to require those behaviors to achieve results - leading to a negative feedback:
Preventing the negative feedback loop in remote work requires thought, investment, and diligence. Engaging employees may be easier in a mature process environment where performance data can help create incentive structures, but employee engagement is never something to be put on autopilot. Even kind environments do not create engagement automatically and can fall victim to the loop.
An engaged worker requires continued investment to maintain engagement. An unengaged worker may ultimately result in turnover. The battle for remote Employee Engagement never ends, but it can be lost.
Design your remote work policies wisely
The factors described above will occur in different mixes at every organization and for every role. Leaders need to assess each situation to determine how to best leverage onsite vs. remote work. The option for remote work may be desirable for both the workforce and employers. However, to get the best results, certain roles need to be encouraged or even required to work from the office. Here are some steps to determine which ones and to achieve overall better performance.
Step 1. Assess your roles for wickedness and process maturity
To develop an effective remote work plan & policy, assess the conditions of the role(s) in question. To help you assess whether your scenario is right for remote work, we offer the following matrix:
Note that low process maturity kills remote suitability. A high number of handoffs, frequent troubleshooting, pervasive vagueness, etc., have an exponentially negative impact on process efficiency in a remote setting.
Be aware that scenarios least suitable for remote work are also those that will have the most risk of low engagement and a negative engagement cycle - invest heavily in these to maintain engagement.
If you have a low-wicked process with low-process maturity, invest in process improvements - these types of investments will pay off more in a remote environment than they did in an onsite environment (see step 4).
Step 2. If more office time is needed, make the change
If an objective assessment of your scenario indicates less than high suitability for remote work, returning to the office may be the answer to efficiency or productivity challenges. However, blanket return-to-the-office policies will meet significant resistance and will usually be unnecessary. Kind, high-maturity work exists and can be done 100% remotely. Some roles may be wicked 100% of the time and would best be done 100% in person. However, as seen in Figure 2, the most common scenarios will be wicked roles that contain some kind tasks or kind roles that contain some wicked tasks, making hybrid remote work models prevalent.
There is a common mistake with hybrid models: letting team members set their own remote days. Even one person being remote is a detriment to collaboration. Here are some guidelines for your hybrid work policies:
Have teams working in wicked environments set a team-wide remote day or time when everyone focuses on completing kind tasks
Have teams working in kind environments set common in-person days to work through issues
Make absolutely sure wicked activities like brainstorming, negotiating, coaching, etc., are planned for days when everyone can co-locate
If you are forced to do wicked tasks remotely, invest in leadership & soft skills to mitigate the issues we’ve discussed and leverage remote best practices
Design your remote work policies thoughtfully and according to the type of work your organization is engaging in. Optimal performance in wicked environments will come from teams that are co-located.
Step 3. Invest in an engagement culture that encourages return to the office
The best possible way to get workers to spend more time in the office is to create an environment where they WANT to spend more time in the office. Developing a culture that encourages this behavior will take time and effort; here are some recommendations:
Design a comfortable workplace design to facilitate interpersonal interaction
Foster a community atmosphere with social activities to build relationships
Offer in-office perks that facilitate health and well-being
Communicate the value of physical presence but encourage feedback
Engagement is a risk factor in every remote work model, and it increases as suitability for remote work decreases (see diagram in step 2), getting people to willingly come to the office has enormous benefits.
Step 4. Invest in process maturity and bring people together to make the changes
Doing kind work remotely can add value like reduced cost, improved employee satisfaction, greater flexibility, etc.; however, many leaders overestimate the maturity of their processes. Investments can make these processes more kind and suitable for remote work, but the actual act of improving typically contains wicked work.
If you find yourself in a position where support processes or other work you would expect to be kind are underperforming, bring workers back to the office to invest in process maturity. Continuous improvement, automation, and reductions in waste, variation, and cost have always been priorities. Now, these investments have additional value because of the added effort in collaboration and reduced risk of a negative engagement cycle. For these investments to work, assess the wickedness of the improvement project and make an informed decision about conducting that project in person or remotely (certain phases of such a project should almost always be conducted with colocation).
Step 5. Develop an onsite onboarding plan for young professionals & new hires
All professionals will face developmental requirements that are impeded by a remote setting. In particular, this will affect new employees. Without the benefits of in-person training & relationship building, remote workers will not come up to speed as quickly or ultimately perform as well as in-office peers.
Provide an extended in-person onboarding & training experience. Note that this means support staff & trainers will also have to be onsite and that onboarding in waves will be most efficient for larger organizations
Even after the onboarding period, consider if new hires should be eligible for remote work until they’ve successfully come up to speed on their role. This will require more experienced team members to also have a rotating onsite presence to support new hires
Plan & manage a mentorship system. Assign new employees mentors when they are hired, and measure & monitor the program to make sure mentorship is established before the new employee starts working remotely
Make sure that resources like knowledge libraries, work instructions, support desks, etc. are all easily accessible remotely, and monitor their usage
The best organizations will lean into colocation for roles that regularly engage in wicked work
Remote work has not been uniformly effective, and not all roles are suitable for remote work. Sam Altman of Open AI recently said this:
“I think definitely one of the tech industry’s worst mistakes in a long time was that everybody could go full remote forever...”, “I would say that the experiment on that is over, and the technology is not yet good enough that people can be full remote forever...” (7)
The reasons why should now be more apparent. To get the best results, employees working on your toughest and most valuable tasks need to spend more time at the office. As a leader, it is your responsibility to get your employees to the office for the right amount of time and at the right times.
(3) Business Insider
(4) Finance Buzz