360 Degree Feedback Process: Guide for Indian SMBs

Most founders and HR heads in Indian startups discover the limits of the annual manager review the hard way. A team lead who looks excellent to the CEO turns out to be a bottleneck for her juniors. A quiet engineer who never speaks in town halls turns out to be the person every other team quietly relies on. The manager's view is one angle, and it is often not the widest one. A well-run 360 degree feedback process fills that gap by collecting structured input from the people who actually work with someone: their manager, peers, direct reports and sometimes internal customers.

But 360 feedback has a reputation problem. In many companies it has become a once-a-year ritual of anonymous comments, awkward conversations and spreadsheets nobody reads. In Indian workplaces, where hierarchy, politeness and "log kya kahenge" shape how people speak up, a badly designed exercise can do more harm than good. A well-designed one, on the other hand, can become the most honest conversation a manager has all year.

This guide is written for founders, HR managers and people leaders at small and mid-sized Indian businesses and startups, typically 20 to 500 employees, who want to run 360 feedback without a dedicated people-analytics team. We will cover when it makes sense, how to design questions and scales, who should rate whom, how to protect anonymity, how to run the cycle step by step, how to handle India-specific cultural dynamics, how to deliver results, whether to link them to ratings and pay, the mistakes that sink most programmes, and a ready-to-use question bank.

What 360-degree feedback actually is (and what it is not)

In a 360 review, a person receives structured feedback from several groups around them instead of only from their reporting manager. The usual groups are:

• Self: the person rates themselves on the same questions.
• Manager: the direct reporting line.
• Peers: colleagues at a similar level, in the same or adjacent teams.
• Direct reports: people who report to the person being reviewed.
• Others (optional): internal customers, cross-functional partners, or, in client-facing roles, selected external stakeholders.

The output is not a score to be argued over. It is a picture of how someone's behaviour lands with different audiences, and, more importantly, where those audiences disagree with each other or with the person's self-view. That gap analysis is where most of the value sits.

A 360 is also not any of the following:

• A replacement for performance appraisal. It measures behaviours and working style, not goal achievement. Targets, OKRs and delivery metrics still belong in your performance review.
• A complaint box. It is not the channel for harassment or ethics concerns. Those need a proper grievance route and, where applicable, the Internal Committee under the POSH Act.
• A popularity contest. If the design rewards being liked rather than being effective, you have built the wrong instrument.
• A one-time event. Value comes from the follow-up: the conversation, the development plan, and the re-measure a year later.

When a 360 degree feedback process makes sense for an SMB or startup

Not every company at every stage needs a 360. Running one too early burns goodwill; running one too late means problems have already calcified. Here is a practical way to judge.

Good signals that you are ready

• You have people managers, and they are new at it. First-time managers, often promoted because they were strong individual contributors, are the single best use case. They rarely get honest feedback on how they lead.
• The team has crossed roughly 25 to 30 people. Below that, the founder usually sees everything directly and honest feedback happens in the corridor. Above it, visibility drops.
• There are cross-functional dependencies. If product, engineering, sales and operations constantly rely on each other, a manager's view of someone is incomplete by definition.
• You are identifying future leaders. Before promoting someone into a leadership role, input from peers and reports is far more informative than a single manager's opinion.
• Attrition clusters around specific managers. A pattern of exits from one team is a prompt to look at upward feedback.
• You have at least a basic culture of trust. People must believe that honest feedback will not be used against them or against others.

Signs you should wait

• The company is in survival mode, with layoffs, delayed salaries or leadership churn. People will not give candid input when they fear the results will feed redundancy decisions.
• There is no manager capable of a good feedback conversation. Collecting input you cannot act on is worse than not collecting it.
• Teams are tiny. With three direct reports, anonymity is practically impossible.
• Leadership wants it as a weapon. If the real intent is to build a case against someone, be honest and use a performance process instead.

Where 360 is the wrong tool

If the problem is an individual's missed targets, a skills gap that is easily measured, or a policy violation, other instruments are faster and fairer. A 360 is best for questions like: how does this person collaborate, communicate, lead, make decisions and handle pressure?

Decide the purpose first: development or evaluation

This is the single most important design decision, and most Indian SMBs skip it. A 360 can serve one of two broad purposes:

• Developmental: results go to the individual (and perhaps their manager or a coach) to help them grow. They are not used to decide pay, promotion or ratings.
• Evaluative: results feed into formal decisions about ratings, bonuses or promotions.

These two purposes pull in opposite directions. When raters know their words may change a colleague's increment, they soften feedback, trade favours or retaliate. When the subject knows the results will affect their pay, they become defensive and lobby raters. Honesty drops in both directions.

For most SMBs and startups starting out, we recommend a developmental first cycle, then, if you want, using the results as one input among several in calibration discussions in later years (more on that below). Whichever you choose, announce the purpose clearly and stick to it. Changing the rules after collecting feedback destroys trust for years.

Designing the questions

Good 360 questions describe observable behaviour, are specific to your context, and can be answered by someone who has actually worked with the person. Poor ones ask for personality judgments or abstract traits.

Principles for writing questions

1. Focus on behaviours, not traits. "Explains the reasoning behind decisions" is observable. "Is a good leader" is not.
2. One idea per question. "Communicates clearly and listens well" is two questions. Raters who see both strengths and weaknesses cannot answer honestly.
3. Use plain language. Many Indian teams work across English proficiency levels. Avoid jargon and idioms like "moves the needle" or "boils the ocean."
4. Tie questions to your values and role expectations. If "ownership" is a company value, ask about ownership in specific forms: follows through, flags risks early, does not wait to be chased.
5. Keep it short. Aim for 15 to 25 rating questions and 2 to 4 open-text questions. Beyond about 15 minutes of effort, quality collapses, especially for raters who have several people to review.
6. Vary the set by role. A manager of people should be asked about coaching and delegation. An individual contributor should not be.
7. Avoid leading or loaded wording. "Does she ever dominate meetings?" invites a particular answer.

A competency framework that works for most SMBs

You do not need a 40-competency model. Five to seven competencies are plenty. A practical starting set:

Choosing the rating scale

A rating scale is easy to get wrong, and the choice affects how much useful variation you get.

Frequency scale versus agreement scale

Two approaches dominate:

• Agreement scale ("Strongly disagree" to "Strongly agree"). Easy to build, but tends to produce uniformly high answers because people agree with positive statements out of politeness.
• Frequency scale ("Never" to "Always"). Asks how often a behaviour is seen. This is generally easier for raters to answer honestly and avoids the awkwardness of "disagreeing" with a colleague.

For Indian workplaces, where raters often hesitate to give anything that feels like a harsh judgment, a frequency scale is usually the better default. Saying "Sometimes" feels less personal than "Disagree."

How many points

• A 5-point scale is familiar and easy to read, but many raters cluster at 4.
• A 4-point scale (no midpoint) forces a lean. It is useful when you notice everyone selecting the neutral middle.
• A 6-point scale adds granularity but can feel fussy.

For a first cycle, a 5-point frequency scale with clear labels is sensible. Always add a "Not enough information to answer" option. Without it, raters guess, and guesses pollute the data.

Open-text questions

Numbers show where; words show why. Include a small set of open questions, and phrase them to produce actionable input:

• "What is one thing this person should keep doing because it helps the team?"
• "What is one thing this person could start or do differently to be more effective?"
• "Is there anything else that would help them grow?"

The "start, stop, continue" structure works well because it forces specificity. Encourage raters to include an example. A comment such as "Needs to communicate better" is nearly useless. "In the last release, I learned about the delay from another team, not from you" is something a person can act on.

Choosing raters

Who gives feedback matters as much as what they are asked. A poorly chosen rater group produces biased or thin data.

How many raters

A reasonable target for each subject:

Total of about 8 to 12 raters per person is a sound range. Fewer than that and a single outlier skews the picture; many more and raters suffer fatigue.

Who picks the raters

There are three common approaches:

• Subject nominates, manager approves. The most common. The person suggests raters, and the manager or HR checks the list for balance. The risk is that people choose friends, so the approval step matters.
• Manager or HR assigns. More objective, but can feel imposed and may miss people who actually work closely with the subject.
• Hybrid. The subject nominates a longer list, and HR selects a balanced subset including at least some people the subject did not name.

We suggest the hybrid. Ask the subject to nominate six to eight people, then have the manager add two or three they know have worked with the person on difficult projects.

Eligibility rules that save headaches

• Raters should have worked with the person for at least three to six months.
• Nobody should be asked to rate more than four or five colleagues in one cycle. Beyond that, feedback quality drops sharply.
• Avoid raters who are in an active conflict or disciplinary matter with the subject, unless their input is handled with care.
• New joiners with under three months of tenure should receive feedback only through a lighter pulse or onboarding check-in, not a full 360.

Anonymity and confidentiality

People give candid feedback when they believe it cannot be traced to them. But anonymity is a promise you must be able to keep, and in small companies that is harder than it sounds.

Practical rules to protect anonymity

1. Set a minimum group size. Do not show results for a rater group with fewer than three respondents. Merge the scores into "Peers and others" or hide that group.
2. Do not break results down by team or seniority if doing so would reveal who wrote what.
3. Keep the manager's feedback identifiable. Manager feedback is normally attributed, because a manager cannot hide behind a group.
4. Edit open comments before sharing. HR or a neutral facilitator should lightly edit comments that identify the writer (for example, "when we were on the Pune client call last Tuesday") or that are abusive or personal.
5. Restrict who can see raw data. Ideally only one or two HR people. The manager sees the consolidated report, not individual responses.
6. Store data securely and decide retention upfront. Tell people how long results are kept and who has access. This matters under the Digital Personal Data Protection Act, 2023, since feedback is personal data about identifiable employees. Have your legal advisor confirm how it applies to your setup.

Honesty about the limits

In a team of eight, a direct report who writes a long, detailed comment may be identifiable by style alone. Be honest with raters: "Your scores will be combined with others and your name will not be shown. Comments are shown without names, but please avoid details that would identify you." Over-promising and then having someone guess a rater's identity is the quickest way to kill the programme.

Anonymous versus confidential

Some companies choose a confidential model rather than fully anonymous: HR knows who said what, but the subject and manager do not. This has an advantage, since HR can follow up if a comment suggests misconduct or a serious problem. The disadvantage is that raters may self-censor knowing HR can see their names. Whichever you pick, state it plainly in the invitation.

Handling cultural hierarchy sensitivities in India

This is where a generic, imported 360 template often fails. Practices designed in flatter, more direct cultures meet a workplace where seniority, age and relationship carry heavy weight. Understanding the dynamics lets you design around them.

Common patterns to expect

• Deference upward. Juniors are reluctant to rate a manager or founder below "good," even when they have real concerns. Rating your boss low can feel disrespectful or risky.
• Inflated, polite scores. Peers often rate generously to preserve relationships. The result is a cluster of 4s and 5s that tells you little.
• Reluctance to give negative written comments. Open text may be filled with praise, while the real message is hidden in what is not said.
• Age versus rank tension. A 27-year-old manager leading a 40-year-old senior engineer creates complex dynamics around who can give feedback to whom.
• Founder and family-business effects. In founder-led or family-run firms, feedback to the promoter can feel off limits.
• Fear of identification. In smaller offices, people fear being identified and quietly punished, even if no one has done so before.
• Different comfort with directness across regions and teams. Teams drawn from different cities, backgrounds and prior employers will vary a lot in how directly they speak.

Design choices that help

Start from the top. The founder and senior leadership should go first, receive feedback, and share what they learned with the team: "Here is what I heard, and here is one thing I will change." Nothing builds trust faster. If leaders exempt themselves, the exercise will be read as a tool for managing people down.

Use behaviour-based, frequency-based questions. "How often does your manager explain the reasoning behind decisions?" feels safer than "Is your manager a good communicator?" It asks the rater to report observations rather than pass a verdict.

Make negative feedback easier to give. Frame questions around growth: "What is one thing that would make this person even more effective?" People find it more comfortable to suggest an improvement than to criticise.

Offer a short, supportive briefing. Spend 20 minutes with raters explaining what good feedback looks like, with examples of useful versus unhelpful comments. Many people simply have never been taught how.

Allow raters to choose their language comfort. If your teams are more comfortable in Hindi or another language for written comments, allow it. HR can translate where needed, preserving meaning rather than polishing it.

Be careful with upward-feedback groups. If a manager has only two direct reports, do not show those scores separately. Consider rolling their feedback into a broader group, or run a short skip-level conversation instead.

Coach managers before the results land. A manager whose authority has rarely been questioned may take upward feedback as an insult. A short briefing on how to receive feedback, listen without defending, and say thank you, is worth the time.

Mind seniority in the debrief. When a senior leader receives critical feedback from juniors, the debrief should be with someone the leader respects, such as the CEO, an external coach or a trusted advisor. A junior HR executive reading difficult comments to a director rarely goes well.

What not to do

• Do not publicly reveal who gave low ratings or hunt for the source of a comment.
• Do not ask raters to justify scores in front of the subject.
• Do not treat low upward-feedback scores as insubordination. If that is how it will be used, cancel the exercise now.

Running the 360 degree feedback process, step by step

Here is a repeatable cycle suitable for a team of 30 to 500 people. For a first run, consider a pilot with the leadership team and a few willing managers, then scale.

Step 1: Define purpose, scope and ground rules (2 weeks before launch)

Write a one-page charter that answers:

• Why are we doing this? (Development, not pay decisions, for example.)
• Who is included? (All people managers? Selected leaders? Everyone?)
• Who sees the results?
• How will anonymity be protected?
• What happens afterwards?

Get leadership alignment on this page. Everything else flows from it.

Step 2: Build the questionnaire (2 weeks before)

Pick five to seven competencies, write 15 to 25 behavioural questions, add 2 to 4 open-text prompts, and choose your scale. Test it with three or four employees to catch confusing wording. Check that each question can be answered by every rater group, or tag questions by group (for example, "coaching" questions only for direct reports and manager).

Step 3: Communicate and brief everyone (1 to 2 weeks before)

Send a clear announcement from the founder or CEO, not just HR. Include the purpose, timeline, how long it takes, confidentiality rules and what is expected. Then hold a short briefing, live or recorded, covering:

• How to give specific, fair, behaviour-focused feedback
• How to avoid bias (recency, halo effect, personal friction)
• What anonymity means
• What happens with the data

Step 4: Select raters (1 week before)

Subjects nominate; managers and HR review and finalise as discussed above. Confirm each rater's workload so no one receives an unreasonable number of requests. Lock the list.

Step 5: Launch and collect responses (2 weeks open)

Open the survey for about 10 to 14 days. Tips:

• Allow raters to save and return to a partially completed form.
• Send reminders at day 3, day 7 and 2 days before close. Reminders should go to non-responders without revealing individual names to managers.
• Keep a target of 80 percent or more completion; follow up personally on gaps where a rater group is below the minimum threshold.
• Make it mobile-friendly. Many staff, especially in field, retail and operations roles, will answer from a phone.

Step 6: Quality-check the data (3 to 5 days)

Before anything goes to anyone:

• Check each rater group has met the minimum number.
• Look for straight-lining (the same score for every question), which suggests careless completion.
• Review open comments for identifying details, abusive language or serious allegations that need to go through a separate process.
• Flag any serious concerns raised in comments (harassment, safety, fraud) and route them to the right channel immediately rather than saving them for the debrief.

Step 7: Generate reports (2 to 3 days)

A good individual report contains:

• A summary of strengths and areas for growth by competency
• Self versus others comparison, with the gaps highlighted
• Scores broken out by rater group (where the group is large enough)
• Top-rated and lowest-rated individual questions
• Selected open-text comments, grouped by theme
• A "blind spots" and "hidden strengths" view: items where you rated yourself higher than others did (blind spots) or lower (hidden strengths)

Resist the urge to make the report fancier than needed. A clear two-to-four page report beats a twenty-page one that nobody finishes.

Step 8: Prepare the debrief (before delivery)

Brief the facilitator, whether HR, a coach or the manager, on the report. They should read the report in advance, find two or three key themes, and prepare questions rather than conclusions.

Step 9: Deliver results in a one-on-one conversation

Covered in detail in the next section.

Step 10: Create a development plan (within 2 weeks of the debrief)

Each person picks one or two focus areas, not seven. They write specific actions, a timeline and a way to check progress. Examples: "Share weekly priorities with my team every Monday" or "Ask for feedback after every sprint review." Their manager supports with time, resources or coaching.

Step 11: Follow up (at 30, 90 and 180 days)

Short check-ins ask: what have you tried, what has changed, what do you need? Some companies run a small pulse survey with the same raters after six months, using only the one or two questions linked to the person's goals. This closes the loop for raters, who see that their feedback led to something.

Step 12: Review the programme itself

After the cycle, ask participants and HR what worked and what did not. Track completion rates, time spent, and how many people completed a development plan. Use that to improve the next round.

Sample timeline

Delivering results so they actually change behaviour

How results are delivered decides whether a 360 becomes growth or resentment. Most people read their report with a mix of curiosity and dread, and the first reaction is often defensive.

Who should deliver

• For most employees, the reporting manager, after HR has briefed them, is a reasonable choice, provided the manager has a trusting relationship and is not the subject of a harsh comment in the report.
• For senior leaders, use an external coach, a board advisor or the CEO.
• For managers who struggle with difficult conversations, HR should sit in or lead the debrief.

Structure of a good debrief (about 60 to 90 minutes)

1. Set the tone. Explain that this is information, not a verdict. Everyone has strengths and blind spots.
2. Let the person read first. Give them 10 to 15 minutes alone with the report, or send it a day earlier. Reading with someone watching makes people guarded.
3. Ask for first reactions. "What stood out? What surprised you? What felt familiar?"
4. Start with strengths. Not as a cushion, but because people need to know what to protect. Strengths also tend to be underused.
5. Move to patterns, not single comments. One harsh comment is noise; a theme across several raters is signal. Focus on themes.
6. Explore the gaps. Where self-rating and others' ratings diverge, ask curious questions. "What might explain why your team rates this lower?"
7. Decide on one or two priorities. Agree on what to work on, not everything.
8. Close with next steps. Date for the development plan, support needed, and when you will meet again.

Language that helps

• "Several people mentioned..." rather than "Your team said..."
• "This is a pattern worth looking at" rather than "This is a problem."
• "What would you like to try?" rather than "You need to fix this."

Language that hurts

• Guessing who wrote a comment.
• Softening the message so much that the person misses it.
• Softening nothing and reading harsh comments aloud.
• Comparing the person to colleagues by name.

When feedback is very negative

Sometimes a report is genuinely painful, for instance a manager whose team uniformly describes fear or disrespect. Slow down. Give the person time, offer a second conversation, and consider coaching support. Do not abandon them with a bad report. At the same time, if comments describe bullying, discrimination or harassment, that goes through the appropriate formal process, not just a development plan.

Share something with the wider team

After the cycle, the leadership team can share aggregate, non-identifying themes: "Across the company, people said they want clearer priorities and more recognition; here is what we will do about it." This shows raters their time was worthwhile and is one of the cheapest trust-building steps available.

Should 360 feedback be linked to ratings and pay?

This is the question every founder asks, and the honest answer is: be very cautious.

The case for keeping it separate

• Honesty drops when stakes rise. Raters hold back or trade favours if they know scores affect compensation.
• Gaming increases. Subjects may lobby for good scores, choose friendly raters or discourage critical feedback.
• Retaliation risk. Managers who see which direct reports gave low scores may treat them differently.
• Reliability. A handful of ratings from a small group is a weak basis for a financial decision about one person.
• It conflicts with development. People cannot simultaneously be open about weaknesses and defend a bonus.

The case for some linkage

• Behaviours matter. A high performer who treats colleagues badly damages the business, and a performance review based only on output can miss that.
• Credibility. If leaders say "how you work matters" but pay ignores it, employees notice.
• Better calibration. Managers can be blind to how someone treats peers and juniors.

A sensible middle path

Many SMBs do best by keeping a clear separation in the early years and then moving gradually toward limited use:

If you move toward linkage, follow these safeguards:

1. Announce the change a full cycle in advance. Never retrofit.
2. Use themes, not raw scores. Look for consistent patterns across raters rather than averages.
3. Require minimum rater numbers for any decision-relevant use.
4. Combine with other evidence. Never let 360 feedback alone drive a promotion or a termination.
5. Apply it to leadership behaviours only, not to every employee.
6. Review for bias. Check whether certain groups, such as women, newer joiners or people from particular backgrounds, receive systematically different feedback. If so, correct before using results in decisions.
7. Document how it was used. In disputes, clarity about process protects both the company and the employee.

Overall, our practical advice: start developmental, build trust, and let behaviours show up in the appraisal conversation through the manager's judgment rather than a formula. Hold the formula for later, if ever.

Common mistakes that sink 360 programmes

Most failed 360s share the same handful of errors.

1. No clear purpose. Raters do not know whether their words affect pay, so they hedge. Fix: write the charter and share it.
2. Too many questions. Fifty items and three open-text boxes per person leads to rushed answers. Fix: keep it under 25 rating items.
3. Too many people to rate. Asking someone to review eight colleagues guarantees shallow input. Fix: cap at four or five.
4. Leaders exempt themselves. Nothing signals hypocrisy faster. Fix: leaders go first.
5. Anonymity that is not real. Small groups, identifiable comments and leaked results end trust. Fix: minimum group sizes and comment editing.
6. Delivering a PDF and walking away. Results emailed without a conversation tend to be ignored or misread. Fix: always hold a debrief.
7. Focusing on weaknesses only. Constant fixing of gaps ignores strengths. Fix: spend half the debrief on what to keep and build.
8. Too many development goals. Seven action items means zero progress. Fix: one or two.
9. No follow-up. Without check-ins, a 360 becomes an annual ritual. Fix: 30-90-180 day touchpoints.
10. Using it to punish. The moment one person is penalised for honest upward feedback they received, the programme is finished. Fix: protect both raters and subjects, and be explicit about it.
11. Copying a foreign template unchanged. Questions written for another culture or industry may not suit your context. Fix: adapt the language and competencies.
12. Treating averages as truth. A 3.8 versus 4.1 difference is usually meaningless. Fix: look at gaps between groups, themes in comments and consistent patterns.
13. Running it too often. Quarterly 360s cause fatigue. Fix: annually, or every 12 to 18 months, with lighter pulse check-ins in between.
14. Ignoring bias. Recency bias, halo effect, similarity bias and gender bias all creep in. Fix: train raters, ask for examples and review results for patterns.

A worked example

Consider a 60-person SaaS company in Bengaluru. Rohit, a strong backend engineer, was promoted to engineering manager nine months ago. His own manager, the CTO, sees fast delivery and rates him highly. Two of Rohit's five reports have recently resigned.

The company runs a developmental 360 with eight raters for Rohit: the CTO, three peers (a product manager, a QA lead and a designer), and four of his direct reports. The 5-point frequency scale gives these results on selected items:

The pattern is clear: Rohit and the CTO see a dependable engineer, but his reports experience a manager who decides alone and gives little feedback. The open comments say things like "He fixes things himself instead of letting us learn" and "I am not sure how my work is evaluated."

In the debrief, run by the CTO with HR support, Rohit's first reaction is defensive: "I only step in because we have deadlines." The CTO responds with curiosity: "What would it take to let someone else own the next module, even if it is slower?" They agree on two priorities: a fortnightly one-to-one with each report that includes feedback, and handing one significant task to a junior with a clear review checkpoint. At 90 days, a three-question pulse with the same four reports shows movement.

Notice what made this work. The data was behavioural, the gap analysis was the centrepiece, the debrief was a conversation, the actions were few, and the follow-up was built in.

Sample question bank

Use this as a menu rather than a script. Pick the items that match your values and role, and adapt the wording. Ask raters to respond using the frequency scale described above.

Communication

• Shares information the team needs in time to act on it.
• Explains the reasons behind decisions, not just the decisions.
• Listens to others without interrupting or dismissing their view.
• Keeps written updates (emails, chat messages, documents) clear and to the point.
• Speaks up respectfully when they disagree.

Collaboration

• Works constructively with people in other teams.
• Shares credit for successes with those who contributed.
• Handles disagreement calmly and looks for a workable solution.
• Offers help to colleagues without being asked.
• Respects others' time and commitments.

Ownership and reliability

• Does what they say they will do, by when they said it.
• Raises risks and delays early rather than at the last minute.
• Takes responsibility for mistakes instead of blaming others.
• Follows through without needing repeated reminders.
• Maintains consistent quality under deadline pressure.

Problem solving and judgment

• Breaks complex problems into clear steps.
• Makes decisions with reasonable speed when information is incomplete.
• Asks good questions before jumping to solutions.
• Learns from past mistakes and applies the learning.
• Backs recommendations with facts or examples.

Leading and developing others (people managers)

• Sets clear expectations and priorities for the team.
• Gives specific, timely feedback, both positive and constructive.
• Delegates meaningful work and trusts the team to deliver.
• Makes it safe for team members to disagree or admit mistakes.
• Recognises good work fairly and visibly.
• Supports team members' learning and career growth.
• Treats every team member fairly, regardless of background, gender or closeness to the manager.
• Is approachable when someone needs help or has a concern.

Adaptability and resilience

• Stays steady and constructive when plans change.
• Responds to feedback with openness rather than defensiveness.
• Learns new tools or approaches quickly.
• Remains respectful when under stress.

Customer or stakeholder focus

• Understands what internal or external customers actually need.
• Responds to requests within a reasonable time.
• Keeps stakeholders informed about progress and changes.
• Takes quality seriously before handing work over.

Open-text questions

• What are the two or three things this person does best that the team should not lose?
• What is one thing they could start doing that would make the biggest positive difference?
• What is one thing they could do less of or stop doing?
• Please share a recent example that illustrates your feedback. (Avoid details that would identify you.)
• Is there anything else you want them to know?

Questions for self-assessment

• Which two strengths do you think others value most in you?
• Which area do you think you most need to improve?
• What kind of support would help you most in the coming year?

Making the process easier with the right tooling

You can run a small pilot with spreadsheets and forms. It gets difficult fast. Matching raters to subjects, protecting anonymity, sending reminders, merging results and producing reports for dozens of people involves a lot of manual effort, and each manual step is a chance to leak data or make an error.

When evaluating a tool or HRMS module for this work, look for:

• Flexible rater selection and approval workflows
• Configurable questionnaires and scales
• Automated anonymity thresholds
• Reminders that do not reveal who has not responded to managers
• Clear individual reports with self-versus-others comparison
• Access controls so only the right people see the data
• A way to attach development actions and follow-ups
• Mobile-friendly forms for non-desk teams
• Integration with your employee records, so reporting lines and rater lists stay current

Frequently asked questions

1. How often should we run a 360 degree feedback process?

Once a year, or once every 12 to 18 months, is enough for most SMBs. More frequent cycles cause fatigue and leave no time for people to act on feedback. In between, use short pulse questions or regular one-to-ones to keep feedback flowing.

2. How many people should rate each employee?

Aim for roughly 8 to 12 raters: the manager, 4 to 6 peers, 3 to 5 direct reports (for managers) and optionally 2 to 3 cross-functional colleagues. Do not report a group's scores if fewer than three people responded in that group.

3. Should we run 360 feedback for every employee or only managers?

For a first cycle, start with leaders and people managers. They benefit most, and their behaviour shapes everyone else's experience. Extending it to individual contributors can come later, ideally with a shorter questionnaire focused on collaboration and communication.

4. Is anonymous feedback better than named feedback?

Anonymous feedback is usually more candid, especially in hierarchical cultures. Keep the manager's input named. In very small teams, where anonymity cannot truly be protected, consider a facilitated conversation or a confidential model in which only HR sees names.

5. What if someone receives very negative feedback?

Treat it with care. Give them time to absorb it, hold a supportive debrief, focus on patterns instead of individual comments, and offer coaching. If comments raise allegations of harassment, discrimination or other serious misconduct, handle those through the proper formal channel in parallel.

6. How do we stop raters from giving everyone high scores?

Use a frequency scale rather than an agreement scale, require examples in comments, brief raters on what useful feedback looks like, and include an option for "not enough information." Focus on gaps between groups and on comments rather than the absolute scores.

7. Can we use 360 results for promotions and increments?

It is better not to, at least at the start. Using results for pay or promotion makes raters guarded and tempts subjects to game the exercise. If you later include it, announce the change well in advance, use patterns rather than raw averages, and combine it with other evidence.

8. How long does a full cycle take?

Plan for about eight to ten weeks from design to debrief for a first cycle, including two weeks of data collection. Subsequent cycles run faster since the questionnaire and process are already in place. Development follow-ups continue for three to six months after that.

Conclusion

A good 360 degree feedback process is not a clever survey. It is a disciplined routine: a clear purpose, behaviour-based questions, a balanced set of raters, anonymity you can truly protect, and a conversation afterwards that leads to one or two concrete changes. In the Indian context, it also requires care for hierarchy, politeness and trust. Start with leaders, keep the first cycle developmental, use frequency scales and growth-oriented questions, and make follow-up non-negotiable.

If you are ready to move beyond spreadsheets and email threads, CozyHR can help you manage employee records, reporting lines, performance cycles and feedback workflows in one place, so that running a thoughtful 360 does not mean a fortnight of manual coordination. You can try CozyHR with your own team and see whether it fits the way you want to run feedback.