We bought a wellbeing app and nothing changed. Why?
Why Individual Interventions Underperform
Most wellbeing spending targets the person's capacity to absorb bad conditions. The evidence favours changing the conditions.
Analysis
Organisations spend heavily on helping individuals cope and lightly on the working conditions that make coping necessary. The evidence does not support that ratio.
What individual interventions do
Resilience training, mindfulness, apps, wellness challenges, individual coaching.
They can produce real benefit, particularly for the people who engage with them, and particularly in the short term.
Effects tend to be modest and to fade without reinforcement.
They do not change the conditions, so the person returns to the same mismatch with a slightly better toolkit.
Why they are popular anyway
Worth being honest, because the reasons are not stupid.
They are fast. A subscription can be launched in a week; reorganising workload takes months.
They are measurable in the wrong way. Sign-up numbers look like success.
They are cheap relative to hiring.
They are uncontroversial. Nobody has to admit that a manager or a staffing decision is the problem.
They locate the problem in the employee, which is comfortable for everyone except the employee.
What the evidence favours
Interventions aimed at the work itself consistently show larger and more durable effects.
Workload and staffing changes.
Increasing control over schedule, method and priorities.
Role clarity.
Manager behaviour, which is a condition of the job for everyone reporting to them.
Reducing interruption and meeting load.
Fixing unfairness, particularly in pay and promotion process.
These are slower, more expensive and more political, which is exactly why they get skipped.
The combination that works
Not either-or. The sensible arrangement is organisational change as the primary intervention, with individual support available for people who want it and clinical routes for people who need them.
The failure is the ratio, not the existence of individual support.
A useful test: if the organisation removed every individual wellbeing benefit tomorrow, would anything about the actual work change? If not, the programme is addressing symptoms only.
The credibility cost
The specific risk of leading with individual interventions.
People notice. A mindfulness session offered to a team that has been short-staffed for a year reads as an insult, and it damages trust in everything the organisation says about wellbeing afterwards.
It sets up a false attribution. When the programme does not work, the conclusion available is that employees did not engage.
It teaches people not to report problems, because the response is predictable and unhelpful.
What to do instead, in order
Ask what the mismatch is, using the six areas.
Fix the biggest one you have authority over, visibly.
Say what you fixed and why.
Then offer individual support, as an addition rather than as the answer.
Sequencing matters more than the content. The same app, offered after a visible workload fix, lands entirely differently.
Where individual support genuinely belongs
The argument against leading with individual interventions is not an argument against having them.
For people with clinical needs, access to care is the employer's contribution and it matters enormously.
For people dealing with life outside work, practical support and flexibility are appropriate and welcome.
For people who want it, coaching and development have value on their own terms.
As a complement to structural change, the same offering lands entirely differently.
The distinguishing question is whether the organisation is also doing the harder thing. Individual support alongside visible change reads as care; individual support instead of change reads as deflection, and employees make that distinction accurately.
The sequencing test
A single question that predicts how an intervention will land.
Has anything about the actual work changed in the last year?
If yes, individual support is heard as an addition and gets used.
If no, the same offering is heard as a deflection and gets ignored, regardless of its quality.
This is why identical programmes succeed in one organisation and fail in another. The variable is not the programme.
Ask the question before launching anything, and if the answer is no, do the structural thing first.