How are you, really?

"How are you?" "Fine."

We ask the most important question in the worst possible way. "How are you?" gets "fine" — even from people who are anything but. The people who are genuinely struggling are often the least able to see the scale of it themselves, and the least likely to put their hand up.

So there's a gap between how people actually are and what anyone around them can see. You can't support what you can't see, and you can't tell whether anything you're doing is helping. The feeling stays vague, and vague is impossible to act on.

Some things are hard to see from the inside.

It's easy to say "yeah, I'm fine" and mean it — even when things have quietly gotten heavier than you've noticed. The people struggling most are often the least able to see the scale of it. A good check-in helps you notice what's hard to see from the inside — using the same questions researchers and clinicians have trusted for decades.

It's easy to say "I'm fine" and mean it. A good check-in helps you see what's hard to notice from the inside.

Mental Health Check-In turns that into something anyone can use on their own, in a few quiet minutes: a vague feeling becomes a clear picture you can actually see, track over time, and do something about.

Real questions that don't feel clinical.

Underneath the calm surface are real research questionnaires — the DASS-21 alongside established ones for resilience, connection, perfectionism, time management, emotional intelligence and coping style, all based on how you answer, never on anything worn or tracked. You work through plainly-worded questions, and it reads them back to you, in ordinary language — what it tends to mean, and what usually helps. Never a diagnosis, never alarmist — a mirror, not a verdict.

Built end to end: the research, the scoring model, the interface, and the SvelteKit build that ships it live.

Where the care shows.

This is a product where the details are ethical, not just aesthetic. Results are delivered gently and clearly. The rarest and hardest states — the moments where someone might be in crisis — are handled with dedicated, carefully-designed care rather than a generic error screen, which is the single feature I'm proudest of. Nothing auto-sends, nothing overclaims, and every result is only ever exactly what it honestly is.

Because a tool like this has to be trustworthy in the way it's built, not just the way it reads, it's clinician-reviewed, audited to WCAG 2.1 AA accessibility, and covered by a suite of automated tests that run green on every single deploy.

The same signal, for a whole organisation.

The same check-in, aggregated and de-identified, becomes a whole-of-population read for an organisation — the picture a traditional employee assistance program can't give you. Those programs typically reach three to four percent of staff, while the real incidence of distress in a workforce sits closer to thirty percent. The workplace view makes that gap visible, confidentially, so support can be aimed where it's genuinely needed rather than where people happen to ask.

The full range, in one product.

This is the "apps" in web · apps · automations: a genuine product — researched, designed, built and shipped by one person, to a standard a sensitive subject demands.