Why fewer than 5% of employees ever use their EAP

Most Indian employers who buy an Employee Assistance Programme are quietly surprised, a year later, by the same number: the EAP utilisation rate sits in the low single digits. A programme covering thousands of people gets used by a few dozen. The vendor calls this normal, and they are right that it is common, but common is not the same as acceptable. When only a small fraction of employees ever pick up the phone, the question is not whether your people are struggling. The evidence on workforce mental health suggests many of them are. The question is why a benefit you are paying for cannot reach them.

The honest answer is uncomfortable. A traditional EAP does not fail by accident. It fails by construction. It is built to wait, to be asked, and to filter people out at every step between distress and help. That design made sense in an earlier era of liability management. It makes far less sense now, when the cost of unaddressed mental health is measurable and the tools to reach people earlier exist. This article is about why the model leaks, how to benchmark your own numbers honestly, and what a proactive first contact actually looks like.

Think about the journey a traditional EAP asks an employee to complete. First, they have to notice they are struggling and name it as a problem worth solving. Then they have to remember the EAP exists, which is hard when it was mentioned once during onboarding and never again. Then they have to find the number, trust that calling it will not get back to their manager, take time out of a shift or a packed day, and make the call. Every one of those steps is a filter. Each filter removes people. By the time you reach the end, you are left with the small minority who were both in acute distress and unusually confident that the system would protect them.

This is why a typical EAP utilisation rate, often reported in the low single digits, tends to be structural rather than circumstantial. The programme is reactive: it sits still until someone in difficulty does all the work of reaching out. The people who most need support, those who are exhausted, withdrawn, or convinced that asking for help marks them as weak, are precisely the people least able to push through that many filters. The design selects against them.

For many Indian workplaces the friction is sharper. Mental health stigma remains strong, the fear that a confidential call could leak to a manager or HR is real, and shift workers and frontline staff often have neither the privacy nor the unstructured time to make a discreet phone call during the day. A programme that depends on a quiet moment and a private phone line is, in effect, designed for a small slice of the workforce.

It is tempting to treat low usage as a culture problem: if only people were more open, they would call. But the EAP model can actively amplify the stigma it claims to address. Reaching out through a standalone helpline is a deliberate, visible act of self-identification. The employee has to decide they are a person with a mental health problem, separate themselves from everyone who seems to be coping, and step into a channel reserved for people in trouble. That framing raises the cost of the first move at exactly the moment a struggling person has the least energy to pay it.

Research on team behaviour points the other way. Amy Edmondson's work on psychological safety, and Google's Project Aristotle study of what makes teams effective, both found that people take interpersonal risks, including admitting they are not fine, mainly when the environment makes those risks feel safe. A once-a-year helpline number does little to build that safety. If anything, the secrecy around it can signal that needing help is something to hide.

There is also a definitional trap worth naming. The WHO's ICD-11 classifies burnout as an occupational phenomenon with three dimensions: exhaustion, cynicism or mental distance from the job, and reduced professional efficacy. Burnout, by that definition, is tied to work conditions, not personal weakness. Yet a reactive EAP can quietly reframe a structural problem as an individual one to be solved by an individual phone call. People sense this mismatch, and many decide, reasonably, that the programme is not built for what they are actually experiencing.

The fix is not a louder helpline. It is to invert the model so that support reaches out before the person has to. Proactive first contact means the system periodically, privately, and without judgement checks in with everyone, not just those already in crisis. A short, anonymised wellbeing check-in normalises the act of reflecting on how you are doing, because everyone receives it. Nobody has to self-identify as the one person who is struggling. The first move becomes ordinary rather than exceptional.

From there, the path to real help should be one step, not five. If a check-in surfaces that someone is having a hard week, the next supportive option, a self-guided exercise, a written resource, or a warm route to a human counsellor, should appear in the same private space, not behind a number they have to dig up and a call they have to nerve themselves to make. The goal is to compress the distance between feeling bad and getting help from a gauntlet down to a single, low-stakes tap.

Two things make this credible in the Indian context rather than intrusive. First, genuine anonymity at the individual level, with HR seeing only aggregate patterns, never names. Second, strict adherence to the Digital Personal Data Protection Act 2023: clear consent, purpose limitation, and respect for the employee's rights as a data principal. Proactive outreach without those guardrails is surveillance, and employees can tell the difference. With them, it becomes what an assistance programme was always supposed to be: support that meets people where they are.

The most common mistake in evaluating an EAP is to confuse provision with impact. Having a programme is not the same as reaching people, and a contract that covers your whole workforce while engaging two percent of it is, functionally, a programme for two percent. The single most useful number you can track is not whether the benefit exists but how many distinct human beings actually engaged with it over the year, and whether that share is rising. Everything else, the polished portal, the hotline brochure, the wellness webinars, is theatre if the reach stays flat.

Benchmark honestly using the tool above, then ask the harder question: is our model reactive or proactive? If it waits to be called, the low number is not a fluke to be fixed with better posters. It is the model working as designed. Platforms built for proactive first contact, including the approach we take at Aha, start from the opposite premise: that the people who need support most are the least likely to ask, so the system should reach them first.

None of this requires believing that an EAP is worthless. It requires being clear-eyed that availability and reach are different things, and that the gap between them is where the human and financial cost quietly accumulates. The figures on lost working days and the return on treating common mental disorders are not arguments for spending more. They are arguments for spending in a way that actually reaches people. A benefit nobody uses is not a benefit. It is a number on a contract, and you already have one of those.