A 120-person Austin fintech, fully remote since 2020, calls everyone within 50 miles back three days a week. Across 47 simulated voices, 40 percent leaned favorable and 50 percent leaned opposed, and reception risk came back high. The verdict is Adjust, and the sticking point is not where the leadership team expected it: the sharpest anger sits with the people hired remote, and what they contest is not the commute. It is a promise.
The floor sits at 15 percent occupancy on an average day, engagement has dipped two quarters in a row, ramp time for new hires keeps stretching, and two senior engineers just resigned citing meeting sprawl and isolation. Sales leadership wants five days on site, engineering managers warn about regretted attrition, and local competitors have announced two or three days. The CEO wants one clear rule for everyone instead of team-by-team exceptions, announced by all-hands email with six weeks notice. One clear rule is exactly what the panel heard, and that is the problem. The same email lands on a junior developer who misses having someone to learn from, on a single dad hired remote who built his childcare around it, on a recruiter who has been selling flexibility for three years, and on the CEO who signed the mandate. The bench puts those readings side by side before the email goes out, not after.
Every voice in the panel, ranked from the most favorable to the most opposed. Read the two ends. The top belongs to executive leadership, at plus 62 on average, and to the local partners who want a busier building. The bottom belongs to the remote-hired employees, at minus 79 on average, with their sharpest voices at minus 90 and below: an engineer with a written remote clause, a compliance specialist who calls it a breach, a single dad who says three days means quitting. The two ends are not arguing about the same thing: one talks culture and collaboration, the other talks about a commitment made in writing. That is why the bench returns Adjust rather than a green light, and why the objection to defuse is the broken promise, not the commute.
Each dot is one voice of the panel, from pushback to support. A lukewarm average can hide a panel cut in two. Here the cloud shows it.
The mandate is read as a breach of the psychological contract for remote-hired employees, who organized their lives around explicit promises of flexibility. Their exit intentions are high, especially among high performers with credible outside options, and their anger is disproportionate to the material change.
Junior developers and office-first roles welcome the mandate as a chance for mentorship and collaboration, but caregivers and long-tenured employees resent the rigidity and lack of carve-outs. Compliance will hinge on whether leadership models attendance and whether the policy feels arbitrary or necessary.
Team leads and HR are caught between enforcing a policy they didn’t design and retaining talent in a hot market. Their buy-in is critical, but the mandate’s blanket application and short notice risk turning them into reluctant messengers of a decision that undermines their authority.
Address the loudest friction first: exempt remote-hired employees from the mandate or offer retention bonuses to those who comply. This signals consistency with past promises and reduces the risk of attrition among high-value talent who feel betrayed.
Middle managers’ skepticism about execution is a shared friction with local partners and a blind spot for leadership. Pilot the mandate with a subset of teams, gather feedback on edge cases (caregivers, commute costs), and adjust the stipend or carve-outs before scaling.
Remote-hired employees (a large share of the panel) are a vocal but low-weight group whose exit won’t sink the policy but will damage employer brand. Craft a separate message acknowledging their contributions, offering flexibility where possible, and framing the mandate as a collaboration boost, not a control measure.
This is an illustrative simulation: a panel of plausible voices generated by an AI from sourced sociological profiles. It is not a poll and not a prediction of what real employees think. Kapari explores the range of possible reactions to help a decision, on a simulated panel, never on the real population.
You describe the decision. A panel of voices reacts. You read the range of reactions before you announce it, not after.