Performance management for a remote team works when you stop managing what you can see and start managing what you agreed, in writing, and reviewed often. Everything else, the dashboards, the activity monitoring, the annual review form, is either a substitute for that or a distraction from it. This guide covers what actually breaks when performance management goes remote, the cadence that fixes most of it, how to measure output without turning into a surveillance operation, and how to document performance so it stands up both in a promotion decision and, in Indonesia, in front of a labour court.
At a glance: Set expectations the employee helped define and can restate without looking them up. Review them weekly in a short one-on-one and formally every quarter. Judge output, not hours logged or green status dots. Write down what you observe as you observe it. That is the whole system, and it works better than what most in-office teams have.
The real problem is not productivity, it is judgment
Start with the finding that should unsettle anyone running a distributed team. Researchers at the University of Warsaw (Anna Matysiak, Agnieszka Kasperska and Ewa Cukrowska-Torzewska) put fictitious employee profiles in front of 937 UK managers who each supervised five or more people, and asked them to make promotion and pay decisions. Published in Work, Employment and Society in June 2025, the results were consistent: with no performance information available, fully remote workers were 11% less likely to be promoted and 9% less likely to be given a pay rise than identical office-based profiles. Hybrid workers were penalised too, at 8% and 7%.
The important part is what happened next. When managers were explicitly told the remote employee performed just as well as the office-based one, the penalty for hybrid workers largely disappeared, but fully remote workers were still 10% less likely to be promoted and 6.5% less likely to get a pay rise. Being demonstrably as good as your in-office peer was not enough to be treated as such.
The field data points the same way. Live Data Technologies' analysis of roughly two million white-collar workers found that just under 4% of fully remote employees were promoted in a year, against 5.6% of those in the office at least some of the time, a gap of about 31%.
Read those two findings together and the conclusion is not that remote work damages careers. It is that unmanaged remote work damages careers, because in the absence of evidence, managers fall back on presence as a proxy for contribution. The fix is not more visibility. It is better evidence, produced deliberately and often enough to be the thing a decision actually rests on.
Annual reviews were already failing. Distance just removes the cover.
The traditional annual review was never carrying its weight. Gallup's research found only 14% of employees strongly agree that the performance reviews they receive inspire them to improve, and that traditional reviews actually make performance worse roughly a third of the time. In a 2024 study, only 2% of Fortune 500 CHROs strongly agreed that their own organisation's performance management system inspires employees to improve. That is the people who own the system saying it does not work.
Two of Gallup's other numbers matter more for remote teams specifically. Only 47% of employees strongly agree they know what is expected of them at work, down from 56% before the pandemic and 61% in 2015. And 56% formally review their goals with a manager once a year or less. An in-office employee with unclear expectations can pick up the missing information by proximity: overheard priorities, a manager's reaction in a meeting, the way a colleague handles something. A remote employee with unclear expectations just guesses, and finds out how they did at the review.
Cadence does more work than any framework you choose
If you change one thing, change how often you talk about performance. Gallup found employees whose manager gives feedback weekly rather than annually are 5.2 times more likely to strongly agree they receive meaningful feedback, 3.2 times more likely to be motivated to do outstanding work, and 2.7 times more likely to be engaged. Eighty percent of employees who received meaningful feedback in the past week are fully engaged. On the formal side, quarterly progress reviews are associated with employees being 90% more likely to be engaged and 2.1 times as likely to say the process is fair and transparent.
A cadence that holds up for a distributed team, in practice:
Weekly one-on-one, 30 minutes, non-negotiable. Not a status update, which belongs in writing. This is for obstacles, priorities and direct feedback on how the work is landing. Cancelling it is the most expensive small decision a remote manager makes.
A short written daily or twice-weekly update from the employee, in one consistent place. It replaces the ambient awareness an office gives you for free, and it doubles as the record you will need later.
Monthly check-in on the whole picture, including workload, wellbeing and whether anything about the arrangement is not working. MixWork runs these for every placement as part of Total Care 360, separately with the employee and with the client, precisely because the two sides rarely raise the same issue first.
Quarterly progress review against goals, written down, with a copy to the employee. This is the artefact that protects them from proximity bias and protects you if performance genuinely deteriorates.
Involve the employee in setting the goals. Gallup found employees who are actively involved in goal setting are twice as likely to have clear expectations. It is also the cheapest way to find out that a target you thought was obvious was not.
Measure output. Monitoring activity makes things worse.
There is a strong temptation, when you cannot see people, to buy software that watches them. The evidence says do not. Research by Chase Thiel and colleagues, published in Harvard Business Review in June 2022, found that monitored employees were more likely to break rules than unmonitored ones, including cheating on a task, taking equipment and deliberately working slowly. The mechanism is the uncomfortable part: being monitored reduced people's sense of responsibility for their own conduct, effectively transferring moral agency to the employer. The researchers also found the effect was weaker where employees felt they were treated fairly.
So keystroke counts, screenshot timers and idle-time alerts do not just annoy good employees, they measure the wrong thing and degrade the thing you actually wanted. What to measure instead:
Deliverables against a date the employee agreed to. Late, on time, or renegotiated in advance, which is a legitimate third option and should be treated as one.
Quality, judged by whoever consumes the work, whether that is a client, a colleague or a downstream team. Written, not implied.
Reliability of communication. Does this person flag a problem early, or does it arrive as a surprise? For a remote role this is a core competency, not a soft skill, and it can be assessed honestly.
Independence over time. The direction of travel matters more than any single month: is this person needing less direction on the same class of work than they did a quarter ago?
A small number of role-specific outcome metrics where the role genuinely has them. Two or three that the employee can influence, not twelve they can only watch.
What all of these have in common is that they are legible to the employee. If someone cannot tell you today whether they are doing well, your measurement system is decorative.
Performance in the AI era: measure judgment, not volume
Output-based measurement needs one update for 2026, because output volume has stopped being a meaningful signal on its own. Microsoft's Work Trend Index 2026, published on 30 June 2026, found that 72% of Indonesian AI users say they can now produce work they could not have produced a year ago, against 58% globally, rising to 82% among the most advanced users. If your metric is volume of output, you will now score everyone highly and learn nothing.
The same report suggests what to measure instead. Indonesia has the highest concentration of what Microsoft calls Frontier Professionals, its most advanced AI users, at 33% against a 16% global average. Those workers named critical thinking as their priority skill at 62% against 46% globally, quality control of AI output at 60% against 50%, and 93% said they treat AI output as a starting point rather than a final answer, against 86% globally. The differentiating skill is not producing more, it is deciding what is good.
That maps directly onto what MixWork sees across its own placements. Every MixWork hire works with AI tools daily as standard, across marketing, operations and engineering alike, and the capability that separates a strong placement from an adequate one is orchestration and judgment rather than raw throughput. MixWork also runs regular professional learning sessions to keep staff current on the AI and software tools their roles actually use, on the reasoning that tooling now moves faster than job descriptions. It is also the direct answer to the objection buyers rarely say out loud about offshore teams, that hires will execute but not think. Judge for judgment and you will find out quickly whether that is true of a given hire.
In Indonesia, the written record is a legal instrument
This is where remote performance management stops being a management topic and becomes a compliance one, and it is the part generic advice on this subject leaves out.
Under PP 35/2021, an employer generally cannot terminate an employee for breaching the employment agreement or company regulations without first issuing successive written warning letters, the surat peringatan sequence known as SP1, SP2 and SP3. Each warning is valid for a maximum of six months unless the employment agreement, company regulations or collective bargaining agreement provide otherwise, and the employee must have a genuine opportunity to respond and improve between them. Termination in Indonesia also follows a defined procedure and, in most cases, attracts statutory severance. Our guide to employee termination in Indonesia covers the grounds and process in detail.
The practical consequence for a remote manager is simple. A performance problem you noticed but never wrote down is not a performance problem you can act on. Undocumented dissatisfaction accumulated over eight months has no legal standing, and attempting to act on it late is how a straightforward underperformance situation turns into a disputed termination. Conversely, a manager who has been running weekly one-on-ones and quarterly written reviews already has the documented trail the process requires, built as a by-product of managing well rather than assembled defensively after the fact.
Hence the rule worth naming, because it holds in both directions:
The Written Record Rule: if a remote employee's performance is not written down, it does not exist, not to your promotion decisions and not to a labour court. The same discipline that protects a good remote employee from proximity bias is what lets you act lawfully when performance genuinely fails.
A framework you can copy
Set up (once per role)
Written role expectations, drafted with the employee, covering what the role owns and what "good" looks like.
Two or three outcome measures the employee can actually influence.
A named point of contact for work direction and a separate one for employment matters, so the employee knows who to raise what with. See our remote onboarding guide for how this splits in practice.
Company regulations and the employment agreement stating clearly what constitutes a breach, before you ever need them.
Weekly
30-minute one-on-one, obstacles and feedback, not status.
Short written update from the employee, in one consistent place.
Monthly
Whole-picture check-in: workload, wellbeing, whether the arrangement is working.
Brief written note of anything notable, positive or negative, while it is fresh.
Quarterly
Written progress review against the agreed expectations, shared with the employee.
Goals revisited and reset jointly.
An explicit promotion and development conversation, so advancement is a scheduled discussion rather than something a remote employee has to initiate against the odds.
When performance genuinely fails
Specific, dated, written feedback with a defined improvement period, before any formal step.
Follow the statutory warning process (SP1, SP2, SP3) with Indonesian legal counsel or your EOR involved from the start, not after the decision is made.
Where MixWork fits
Most of the above is management practice a client owns, and should. What MixWork removes is the part where remote performance management collides with Indonesian employment law and with nobody having time to run the human layer.
Total Care 360 is included by default with the EOR, not metered as an add-on. A dedicated HR manager backed by a full HR team runs the monthly check-ins with the employee and, separately, with the client, so the two accounts of how things are going get compared while the gap is still small. When performance genuinely needs a formal process, the warning sequence and documentation are handled by the legal employer as routine practice rather than improvised under pressure. And the placement profile does a lot of the work up front: MixWork hires average around six years of professional experience, typically built inside multinationals and global agencies, which is the band where someone delivers independently without needing to be supervised into it. Twelve-month retention runs above 90%, which is well ahead of the churn most buyers have experienced from conventional seat-based arrangements.
If you are running or building a remote team in Indonesia and want the performance layer to be structured rather than improvised, talk to us.
Disclaimer: This article is general information, current as of August 2026, and refers to Indonesia's Job Creation Law framework and PP 35/2021 at a summary level. Warning-letter requirements, termination grounds and procedure are fact-specific and subject to change, and company regulations or a collective bargaining agreement may alter them. Always confirm your obligations with qualified Indonesian legal counsel before acting on a performance or termination matter. MixWork provides EOR, HR, and managed IT services, not legal advice.
