# Employee Survey Questions: A Bank by Topic, With Scales

A bank of employee survey questions by topic: engagement, manager, role clarity, workload, inclusion and eNPS, with wording rules and anonymity thresholds.

Published 2026-10-27 · Updated 2026-10-27
Canonical: https://hercules.works/blogs/employee-survey-questions-india

Good employee survey questions are specific enough to act on. Ask about the last two weeks, not about "generally". Ask about a behaviour you can change, such as whether priorities were clear, rather than about a feeling you cannot. Keep the scale identical across waves, keep answers anonymous above a minimum group size, and add one open question per section. The bank below groups 40 questions by topic so you can cut it to the 12 that fit the decision you need to make.

## Start With the Decision, Not the Question

An employee survey that asks everything gives you a report nobody can act on. Name the decision first: whether to change how priorities are set, whether to fix a manager, whether to keep a hybrid policy. Then pick the questions that would change that decision. Twelve focused questions beat sixty generic ones, and a shorter survey raises completion on a phone. This bank is organised by topic so you can cut it down. If you are new to the process, read the [employee survey guide](/blogs/employee-survey-india/) first, then use this page for the wording.

## Engagement and Motivation Questions

Ask about intent and conditions, not about happiness. These eight items work as a core block:

| # | Question | Scale |
|---|---|---|
| 1 | I would recommend this company as a place to work. | Five-point agree to disagree |
| 2 | My work gives me a chance to do what I do best. | Five-point agree to disagree |
| 3 | I intend to still be working here in twelve months. | Five-point agree to disagree |
| 4 | I can see how my work contributes to the company's goals. | Five-point agree to disagree |
| 5 | I have the materials and equipment I need to do my work well. | Five-point agree to disagree |
| 6 | In the last seven days, I felt recognised for good work. | Frequency scale |
| 7 | My work has become more interesting in the last six months. | Five-point agree to disagree |
| 8 | I would recommend this company to a friend looking for a job. | Five-point agree to disagree |

Items 1 and 8 look similar and are not. One asks about the company as an employer, the other about a personal recommendation. Keep both only if you plan to compare them.

## Manager and Role Clarity Questions

Manager items predict more variance than almost anything else you can ask, and they are the easiest to act on. Keep them behavioural.

1. My manager has given me useful feedback in the last month.
2. I know what is expected of me at work.
3. My manager helps me prioritise when everything is urgent.
4. I can raise a problem with my manager without it being held against me.
5. My manager explains the reasons behind decisions that affect my work.
6. I have a conversation about my development at least once a quarter.
7. When I do good work, my manager notices.

Avoid "My manager is good". Nobody can act on it, and the answer depends on the last interaction the respondent remembers.

## Workload, Resources and Wellbeing Questions

Wellbeing questions are sensitive. Ask about conditions and load, not about diagnoses, and never ask a person to disclose a health condition.

1. My workload is manageable in a normal week.
2. I can take breaks during the working day.
3. I have had to work outside my normal hours more than once in the last month.
4. I know where to get support if work is affecting my health.
5. My team has enough people to do the work we are given.
6. I can turn down a request when I am at capacity.
7. My work has been distributed fairly in the last month.

Where you plan to report results by team, check the group size first. A team of five cannot be reported separately and stay anonymous.

## Inclusion and Voice Questions

1. I feel comfortable being myself at work.
2. My ideas are taken seriously, whoever raises them.
3. I have seen a decision change because of employee feedback.
4. People on my team treat each other with respect.
5. I know how to report behaviour that is not acceptable.
6. I believe the report would be handled fairly.

Ask these of the whole company, not of one group, and report the distribution. If you collect demographic data to compare groups, keep every reported cell above your anonymity threshold and drop a cut that fails it.

Workplace diversity survey questions belong here rather than in a separate block. Ask about experience and policy, such as whether a report would be handled fairly, rather than asking people to disclose a category you will not report on. A question that collects a category and then hides the result costs trust and returns nothing.

## eNPS: One Question and Two Follow-Ups

The employee Net Promoter question is standard: "On a scale of 0 to 10, how likely are you to recommend this company as a place to work?" then "What is the main reason for your score?" as an open question.

Calculate eNPS as the share of promoters (9 or 10) minus the share of detractors (0 to 6), leaving passives (7 and 8) out of both. Report the score with its base, because a score from 40 responses moves by several points from noise alone. If you run waves, keep the question and the scale identical or the trend means nothing.

Pair the score with the reason text. A score without the open answer tells you that people are unhappy and nothing about why.

## Wording Rules That Survive Translation

Indian teams often answer in more than one language, and a survey written in English is not automatically understood the same way by everyone. Five rules help:

- One idea per item. Split anything joined by "and".
- Named time frames. "In the last month" beats "recently".
- No idioms. "Go the extra mile" and "firefighting" translate badly.
- No internal jargon. If a term needs induction to understand, drop it.
- One scale per section. Mixing frequency and agreement scales on one screen causes misclicks.

Back-translate any translated version and have a native speaker read it on a phone. The [survey bias guide](/blogs/survey-bias-india/) explains the errors this catches, and the [question types guide](/blogs/survey-question-types-india/) covers the scale design.

## How Many Responses You Need

At 95 percent confidence, a 50 percent metric needs about 400 responses for a margin near plus or minus 5 points, and about 1,000 for plus or minus 3. A 2 point shift between two waves needs roughly 4,800 responses per wave to read with confidence, which is why most companies should treat small wave-to-wave movements as noise.

For a company of 200 people, a census is better than a sample: invite everyone and report the response rate. For a company of 5,000, a sample is fine, but every reported team cut still needs its own base. The [sample size guide](/blogs/sample-size-calculator-india/) has the tables.

## Running This Survey on Hercules

Build the block you need, add routing so a section only shows to the right people, and run a pilot before the full wave. The free plan includes 30 responses, which is enough to test the wording and the scale. Full pricing is [per response](/pricing): ₹150 on Plus from 100 responses, ₹120 on Pro in bundles of 400 to 1,000.

Keep the survey anonymous: do not collect employee identifiers alongside answers, and drop any reported cell below the group size you promised. For company-wide studies the same [pulse survey](/blogs/pulse-survey-india/) cadence applies, with the same questions, so the trend is readable.

## Frequently asked questions

### What questions should an employee survey ask?

Ask about the conditions you can change: clarity of expectations, feedback from the manager, workload, resources, recognition and voice. Add eNPS if you need one headline number. Keep the set short enough to finish on a phone.

### How many employee survey questions is too many?

Twelve to twenty is a practical range for a quarterly survey. Every extra item lowers completion and adds a table nobody reads. If you need more topics, rotate blocks across waves instead of asking everything every time.

### Should employee surveys be anonymous?

Yes for anything about management, workload or culture. Do not store employee identifiers with answers, and do not report a team cut below your stated minimum group size. If you must track change over time, use a self-generated code rather than a name.

### How is eNPS calculated?

Share of promoters, who answer 9 or 10, minus the share of detractors, who answer 0 to 6. Passives, 7 and 8, are left out of both. Report the base with the score, because small bases make the number move on noise.

### Who should be asked about the manager?

Everyone who reports to that manager, including people who joined in the last three months if they have had a feedback conversation. Report a manager's result only when the group is large enough to stay anonymous.

### Can I add open-ended questions to an employee survey?

Yes, one per section at most. Ask for the reason behind a score or one improvement, not for an essay. On a phone, a single short box gets answered; three long boxes do not.