Most employee wellness surveys fail at the same place. People answer honestly, the results go into a slide, and nothing changes. The next survey gets fewer responses, and by the third nobody believes it matters.
This guide is the practical version: 40 questions you can use, how to run them so people trust the answers, what the results can and cannot tell you, and what to do in the 90 days afterwards.
Start with what you will do with the answer
Before you write a question, decide what you would change if the answer came back bad. If nothing would change, leave the question out. Three tests:
- Can you act on it? A question about commutes is only useful if transport is something you can influence.
- Will people believe the answer is safe? If a manager can see who said what, the answers are not real data.
- Can you ask it again in 90 days? A survey is only worth running if you can compare it to the last one.
The six areas worth asking about
These cover what a health agent or a wellness programme can actually help with, without turning the survey into a medical exam. Keep every rating on the same five-point scale, from 1 (poor) to 5 (excellent), and always include a "prefer not to say" option on health questions.
Physical health and movement
- How would you rate your overall physical health today? (1 to 5)
- In a normal week, how many days do you do at least 30 minutes of activity that raises your heart rate? (0, 1 to 2, 3 to 4, 5 or more)
- How many hours a day do you spend sitting at work? (under 2, 2 to 4, 5 to 7, 8 or more)
- Does physical pain (back, neck, shoulders or joints) affect your work? (never, occasionally, most weeks, most days)
- How often do you eat a proper meal during your working day? (never, some days, most days, every day)
- Is there anything about your physical health you would like help with? (open)
Sleep and energy
- On a normal work night, how many hours do you sleep? (under 5, 5 to 6, 6 to 7, 7 to 8, more than 8)
- How often do you wake up feeling rested? (never, some mornings, most mornings, every morning)
- How often does tiredness make it hard to concentrate at work? (never, some days, most days, every day)
- How often do you work outside your contracted hours? (never, a few times a month, weekly, most days)
- How would you rate your energy through the working day? (1 to 5)
- What would help you sleep better or manage your energy? (open)
Mental health and stress
These questions are sensitive. Tell people before they start what will happen with the answers, and give them a way to skip any question without losing their response.
- In the last four weeks, how often have you felt stressed at work? (never, some days, most days, every day)
- How would you rate your mental health today? (1 to 5, prefer not to say)
- How often do you find it hard to switch off from work? (never, some evenings, most evenings, every evening)
- In the last four weeks, how often have you felt low, flat or hopeless? (never, some days, most days, every day, prefer not to say)
- Do you know where to get confidential mental health support if you needed it? (yes, not sure, no)
- Would you use a private chat with a health professional if your employer offered it? (yes, maybe, no)
- How comfortable would you be telling your manager you were struggling? (1 to 5)
- Is there anything about your mental health or stress you would like support with? (open, optional)
Preventive care and screening
- When did you last have a general health check? (under a year ago, 1 to 2 years ago, more than 2 years ago, never)
- Do you know which preventive screenings you are due for your age and sex? (yes, partly, no)
- Have you had your blood pressure checked in the last 12 months? (yes, no, not sure)
- Do you know your HIV status? (yes, prefer not to say, no)
- Is cost a reason you have skipped a health check or a treatment? (never, sometimes, often)
- What would make it easier for you to get a check-up? (open)
Benefits and support awareness
This is where most employers find the cheapest win. People cannot use a benefit they do not know about.
- Do you know what your medical scheme covers? (yes, partly, no, I am not on a scheme)
- Do you know what counselling or employee assistance your employer offers? (yes, I have heard of it, no)
- Have you used any wellbeing benefit in the last 12 months? (yes, no, I am not sure what is available)
- If you needed help, would you know who to ask at work? (yes, not sure, no)
- How easy is it to take time off for an appointment? (1 to 5)
- What benefit would make the biggest difference to you? (open)
- Would you use a free, private health service on WhatsApp? (yes, maybe, no)
Work and environment
- How manageable is your workload most weeks? (1 to 5)
- How often do you take your full lunch break? (never, some days, most days, every day)
- Do you feel able to raise a problem with your manager? (1 to 5)
- How often does work interfere with your family or personal time? (never, some weeks, most weeks, every week)
- Do you have a private place to take a health call during the day? (yes, sometimes, no)
- Is there anything about how work is organised that affects your health? (open)
- What one change would make this a healthier place to work? (open)
How to run it so people answer honestly
- Keep it short. Use 20 to 25 questions for a regular check-in. Use the full 40 for a first baseline. Long surveys get abandoned, and an abandoned survey is worse than a short one.
- Make it anonymous, not just confidential. Do not ask for a name, an email address or an employee number. If your tool needs a login, say in the introduction that responses are reported as totals only.
- Suppress small groups. Never report a result for a group smaller than 10 people, and never split results by manager. A department of four is not anonymous.
- Say what happens next. Give a date when you will share the results, and another for the changes. Two sentences is enough.
- Meet people where they are. Many employees do not sit at a desk. A QR code in the staff room, a WhatsApp link and a paper form reach more people than an emailed spreadsheet.
- Do not use it as a performance review. If results can be used against a team, the next survey is worthless.
POPIA and health information
Under POPIA, health information is special personal information. That means you need a clear purpose and a lawful basis (usually consent), you must secure the data, and you should not keep it longer than you need. Say in the introduction what you are collecting, why, who sees the totals, how long you keep the responses, and how someone withdraws. If you are not sure, ask your legal adviser before you collect anything. This guide is general information, not legal advice.
How to read the results without over-claiming
- Compare yourself to yourself. Your own last survey is the only benchmark you can trust. Be careful with industry benchmarks you cannot verify.
- Look for the lowest scores and the widest gaps. A low score in one area is a place to start. The same question scoring badly in one group and well in another tells you where to look.
- Separate what people say from what people do. "I would use a health service on WhatsApp" is intent. Usage after you launch it is the real measure.
- Do not over-read one round. One survey is a snapshot. Two in a row is a trend.
What to do in the next 90 days
- Publish the top three themes and what you will do about each, with a date. Include the things you are not going to fix, and say why.
- Fix one thing properly. A benefit more people know about beats a new benefit nobody uses. If awareness of counselling is low, that is a communication problem, not a budget problem.
- Make help private and easy to reach. People are more likely to ask for help on a phone, in a chat, than in a room with their manager.
- Run it again in 90 days with the same questions, and show the change.
Where Ollie fits
Ollie is a health agent your people use on WhatsApp, the web or the phone. It answers everyday health questions, runs a 30-second face scan from a phone camera, explains what a medical scheme covers, and books a session with a human expert when someone needs to talk to a person. It is private, and the employer view shows totals, never what anyone wrote.
It is useful after a survey because it closes the loop on the things people ask for: somewhere private to ask a question, help understanding a benefit, and a way to get a check-up without taking a day off. Ollie is a wellness product. It does not diagnose, and it is not a substitute for a doctor.
Questions employers ask
How many questions should an employee wellness survey have?
20 to 25 for a regular check-in, and up to 40 for a first baseline. Completion matters more than coverage: a survey most people finish beats a longer one most people abandon.
Should an employee wellness survey be anonymous?
Yes, if you want honest answers. Do not collect names or employee numbers, report only totals, and suppress groups smaller than 10 people.
How often should we run it?
A baseline, then every 90 days with the same questions. Twice a year is enough if you act on the results each time.
What response rate is useful?
Enough responses to compare groups of 10 or more. If only a small share of the team answers, treat the results as a set of themes rather than a measurement, and say so.
Can we use the results in performance reviews?
No. Using health information against a person or a team breaks the trust the survey depends on, and health information carries special protection under POPIA.
What is the cheapest thing to fix first?
Awareness. Most teams offer more support than their people know about. Fix the communication before you buy anything new.
See how it fits your team
Ollie for teams is a private health agent for every employee, on WhatsApp, web and phone. From R29 per person a month after a 14-day free trial, no card to start.
Or book a 30-minute call and we will scope it to your headcount.
Related: employee benefits in South Africa, and how Ollie works for a team.