Most Indian companies already run a wellbeing survey. Almost none of them can answer the one question that predicts whether people will still be there next year, which is whether the team gets to switch off between working days.
To measure whether a team is recovering, ask a small number of validated questions about rest and detachment every month, at team level, with individual answers never visible to the employer. The WHO-5 Well-Being Index is free, takes two minutes, and includes an item on waking rested. Compare a baseline against a later reading rather than reporting a single number.
The rest of this piece is how to do that without producing a number nobody believes.
Why Doesn't Your Engagement Survey Already Show This?
Because it is measuring a different thing on a different clock. Engagement asks how people feel about the organisation: whether they would recommend it, whether they see a future here, whether their manager supports them. Useful questions. They tell you about attitude toward the employer.
Recovery asks whether a person's system gets to stand down between shifts. It is also the thing that breaks first, which is why we argue it is a better thing to measure than burnout. That is a physiological and psychological state, and it moves fast. In their 2015 review in the Journal of Organizational Behavior, Sabine Sonnentag and Charlotte Fritz summarise daily diary studies where detachment on a given evening predicts how a person feels by bedtime and the next morning, and week level studies where detachment across a week predicts affect by Friday.
An annual survey cannot see something that changes on a Tuesday. Running one and concluding your people are fine is the measurement equivalent of taking a photograph of a river and reporting the depth.
What Should You Actually Be Measuring?
Four things, and none of them require a platform.
Whether people wake rested. The single most direct indicator that the night did its job. If this is low across a team, nothing else in your wellbeing budget is going to matter much.
Whether the worry switches off. Psychological detachment is the technical term, and it is the mechanism by which workload becomes exhaustion. Sonnentag and Fritz report that job stressors, workload above all, predict low detachment, and that low detachment then predicts strain and poorer wellbeing.
Whether people are drained by the end of the day. Distinct from the above, because someone can detach fine in the evening and still be running at a deficit by 6pm.
What the working pattern actually looks like. Not a feeling. A count. When does the last work message of the day get sent. How many meetings sit after 5pm. How often does someone start the day already behind. Self logged for a week by the team is enough, and it is usually the part that changes the conversation, because it makes the pattern visible to the people producing it.
A team that scores badly on rest and well on engagement is not a contradiction. It is a group of people who like where they work and are being worn out by it.
Which Instruments Are Free and Validated?
More than most vendors will tell you, because the free ones do not come with a subscription.
The WHO-5 Well-Being Index. Five items, about two minutes, answered for the past two weeks. It was developed in the 1990s by the late Per Bech at the Psychiatric Centre North Zealand, and in 2024 the copyright was assigned to the World Health Organization, which now publishes it as an open access document under a Creative Commons licence. Scoring is simple: each item runs 0 to 5, total 0 to 25, multiplied by four for a percentage. WHO notes that a score below 50 percent has been suggested as a cut-off indicating further assessment may be needed.
Item four of the WHO-5 asks whether the person woke up feeling fresh and rested. That one line is most of what this whole article is about, and it has been sitting in a free WHO instrument the entire time.
One caution worth reading properly. The WHO-5 is published under a NonCommercial Creative Commons licence. If you are a vendor building it into a paid product, that is a licensing question for WHO rather than an assumption to make, and the document points commercial queries to WHO directly.
The Recovery Experience Questionnaire. Developed by Sonnentag and Fritz in 2007, it is the measure most of the detachment literature uses, and it separates detachment from relaxation, mastery and control. Their 2015 review notes that most studies in this field assess detachment using it. It is a research instrument rather than a WHO publication, so check the terms before deploying it commercially.
A note on the clinical screens. GAD-7 and PHQ-9 are widely used, brief and well validated, and they are screens for anxiety and depression rather than measures of recovery. An employer collecting them is collecting clinical data about named employees, which is a different undertaking with different obligations. If you use them, they should sit behind a clinician, and the employer should never see them.
How Often Should You Run It?
Monthly, and expect the first two readings to be noise.
The evidence on detachment operates at day and week level, so a monthly reading is already a summary rather than a live signal. Annual is useless for this purpose. Weekly usually produces survey fatigue and a falling response rate, which corrupts the data faster than the extra frequency helps.
Give it a baseline before you change anything, then a reading at 30 and 45 days after a change. Anything shorter and you are measuring the novelty of being asked.
What Has to Be True for People to Answer Honestly?
Four conditions, and if you cannot meet all four, do not run the survey at all. A survey people lie on is worse than no survey, because it produces a number that gets quoted in a board deck for the next year.
Individual responses are never visible to the employer. Not to HR, not to the manager, not in a spreadsheet somebody could open. Aggregate only.
A minimum group size. We use eight. Below that, a team is small enough that people can be identified by inference, and everyone in the room knows it.
People get something back on day one. If an individual completes a WHO-5, they should see their own result, privately, with a plain explanation of what it means. Otherwise you are extracting data from people and returning nothing, which they notice.
There is a private route to help. Somebody will score low. That is not a hypothetical, it is arithmetic. There has to be a way for that person to reach an actual clinician without going through their employer to do it.
On the legal side, India now has an operative data protection regime. The Digital Personal Data Protection Act, 2023 is supported by the DPDP Rules notified on 13 November 2025, with compliance phased through to 2027. We are not lawyers and this is not legal advice. Take your own before you collect anything about employee health, and design for aggregate reporting from the start rather than retrofitting it.
What Makes a Result Believable?
A number before, a number after, and a willingness to report the readings that did not move.
Report change over a stated period for a stated group, and stop there. Saying that a team's rested score moved from 2.2 to 2.9 over 45 days while three working pattern changes were in place is honest. Saying that your programme caused it is not, because you have no control group and neither does anyone else selling this. The evidence base for wellbeing offerings aimed at individuals is weak enough that overclaiming is the fastest way to lose a room: Fleming (2024) compared 46,336 workers and found participants no better off than people who did not take part. We have written about what that does and does not show.
The last thing, and it is the one most measurement programmes fail on. Measuring does not fix anything. If you run a diagnostic, publish a number, and change nothing about the working pattern, response rates fall by half the next round and you have taught your team that being asked is a substitute for being heard.
The number is only worth collecting if somebody with authority over the calendar is willing to act on it. If that person is not in the room when you design the survey, you are not measuring recovery. You are documenting it.
What Else Do People Ask About Measuring Team Wellbeing?
Is the WHO-5 free to use?
The World Health Organization publishes the WHO-5 Well-Being Index as an open access document under a Creative Commons Attribution NonCommercial ShareAlike 3.0 IGO licence, following the 2024 assignment of copyright to WHO. Non-commercial use with attribution is permitted under those terms. For commercial use, the document directs queries to WHO's rights and licensing team.
What is the difference between engagement and wellbeing measurement?
Engagement measures how people feel about their employer and is usually surveyed annually. Recovery and wellbeing measures capture whether a person's system gets to stand down between working days, and change over days and weeks. A team can score well on engagement while scoring badly on rest, which is a common and easily missed pattern.
How small is too small to report a team's wellbeing score?
Below about eight respondents, individuals become identifiable by inference even in an aggregate report, and people answer accordingly. A minimum group size of eight is a reasonable floor. Where a team is smaller, either combine it with another group for reporting or do not report on it separately.
Should employers use GAD-7 or PHQ-9 with staff?
Only with care and never with the employer seeing results. GAD-7 and PHQ-9 are clinical screening tools for anxiety and depression, not workplace wellbeing measures. If they are used, they should be administered and interpreted by a clinician, with the employer receiving nothing at individual level and ideally nothing at all.
How long before a change in working patterns shows up in the numbers?
Recovery measures move faster than most organisational metrics because detachment operates at the level of days and weeks. A baseline followed by readings at 30 and 45 days is usually enough to see whether anything shifted. Expect noise in early readings and avoid drawing conclusions from a single measurement.
A note on the cover image
The image at the top of this piece was generated by AI, to a brief written by us. It is not a photograph and does not depict a real person or place.
Sources
World Health Organization (2024). The World Health Organization-Five Well-Being Index (WHO-5). WHO/UCN/MSD/MHE/2024.1. Licence CC BY-NC-SA 3.0 IGO. Full document
Topp, C. W., Østergaard, S. D., Søndergaard, S., & Bech, P. (2015). The WHO-5 Well-Being Index: a systematic review of the literature. Psychotherapy and Psychosomatics, 84(3), 167-176. Karger
Sonnentag, S., & Fritz, C. (2015). Recovery from job stress: The stressor-detachment model as an integrative framework. Journal of Organizational Behavior, 36, S72-S103. Full text
Fleming, W. J. (2024). Employee well-being outcomes from individual-level mental health interventions: Cross-sectional evidence from the United Kingdom. Industrial Relations Journal, 55, 162-182. UK sample. Wellbeing Research Centre, University of Oxford
On the Indian data protection position, professional commentary rather than the statute: EY India on the DPDP Act, 2023 and DPDP Rules, 2025. Take independent legal advice.
