We Spent a Year Saying 'Not Another Just Breathe App'. Then We Shipped Breathing Exercises.

For most of this app's existence, the first line of our positioning has been that ILTY is not another "just breathe" app. We have an entire post arguing that breathing instructions are what mental health software offers when it has nothing to say to you.

In version 1.8.5 we shipped a Toolkit with nineteen practices, and several of them are breathing exercises. Box breathing is in there. So is the physiological sigh.

Somebody is going to notice, so here is the whole argument, including the part where we were wrong.

What we were actually objecting to

Reread the original complaint and it was never "breathing does not work". It was narrower than that, and the narrowness is the point.

The objection was to breathing as a substitute for engagement. You open an app because something is wrong. The app does not ask what. It offers a four-second inhale. The exercise is not a response to your situation, because the app has not established what your situation is; it is the thing it was always going to offer. It would have offered it to anyone.

That is the actual failure, and it is a failure of sequence rather than of technique. The instruction might be fine. Arriving before the conversation is what makes it feel dismissive, and it is why so many people describe these apps as hollow. Being handed a generic action by something that has not asked you a single question is a specific and recognisable experience of not being taken seriously.

What changed our minds

Two things, and the second was more uncomfortable than the first.

The evidence for slow-exhale breathing is genuinely good. Not "wellness good". A 2023 randomised controlled trial run at Stanford compared cyclic sighing, box breathing, and mindfulness meditation over a month of daily practice. All three helped. Cyclic sighing, which is a double inhale followed by a long exhale, produced the largest improvement in mood and the biggest reduction in respiratory rate. That is a specific, replicable, mechanistically sensible finding, and it would be dishonest to keep implying the entire category is theatre because a lot of apps deploy it badly.

Our own users kept describing a state where talking was not available. This is the part that actually moved us. People told us about moments where they could not construct a sentence: the middle of a panic wave, three in the morning, the ninety seconds after a message landed badly. In those moments a companion asking a thoughtful question is not a better product. It is a worse one, because it demands something the person cannot currently produce.

We had built a product with exactly one mode of interaction, and we had been treating that as a principled stand. It was partly a principled stand and partly a gap.

What we actually built

The Toolkit is nineteen practices grouped into four situations: panic waves, anger spikes, rumination, and general regulation. It includes breathing techniques (box breathing, the physiological sigh), grounding (5-4-3-2-1 and variants), and body-based practices.

Three design decisions did the work of keeping this from becoming the thing we criticised:

It is not the front door. The Toolkit is somewhere you go, not something that greets you. Nothing offers you a breathing exercise before you have said anything. The default remains a conversation, because the default is the position the product takes about what you need.

It is offline. Every practice works with no signal. This turned out to matter more than we expected, and it is the subject of its own post: the moments people need this most are disproportionately moments of bad connectivity. A regulation tool that requires a network is not a regulation tool, it is a regulation tool most of the time.

It is nineteen, not two hundred. Curation was deliberate. A library of two hundred exercises solves the app store screenshot problem and creates a browsing problem for someone in distress. Nineteen is small enough to know.

The part where we were wrong

Here is the honest accounting, since we ask other companies for theirs.

We over-rotated on the positioning. "Not another breathe app" is a good line, it is true about most of the category, and we let it harden into a product constraint it was never meant to be. For a while, the case against building any regulation tool was mostly that it would complicate our marketing. That is a bad reason to leave a gap in a product people use at three in the morning.

We conflated a technique with its deployment. Box breathing is not the problem. Box breathing offered by software that has not asked you anything is the problem. Those are separable and we were not separating them, in public, for about a year.

We were slower than we should have been. People described the no-words state to us repeatedly before we built for it. The feedback was clear and we spent months arguing with it internally because it did not fit the story.

What has not changed

The positioning is intact, and it is intact because it was never really about breathing.

ILTY still opens with a conversation. Our companions still push back, and Mr. Relentless will still tell you something you did not want to hear rather than offering to help you regulate through it. We still do not publish clinical outcome claims, because we have not run the studies that would justify them. We still will not tell you it gets better.

What changed is that we now have something for the moments when talking is not on the table. That is an addition to the argument, not a retraction of it.

If we had shipped nineteen breathing exercises instead of the confrontational companion, that would have been the reversal. We shipped them behind it.

Frequently asked questions

Does ILTY have breathing exercises now? Yes. Nineteen practices across four situations (panic, anger, rumination, general regulation), including box breathing and the physiological sigh, all available offline. They live in the Toolkit, not on the home screen.

Is ILTY a meditation app now? No. We also shipped seven guided meditations in 1.8.5, which makes this a fair question. The distinction: meditation and the Toolkit are things you can reach for, and the product's centre is still a conversation that does not agree with you by default. If you want a dedicated meditation app, our tested roundup does not rank ILTY first, because it should not.

Why did it take so long? Some of it was engineering, mostly offline support and seven languages of audio. Most of it was us arguing with our own positioning for longer than the evidence justified.

Does the breathing stuff actually work? Slow-exhale breathing has decent trial evidence for short-term mood and physiological arousal, with cyclic sighing performing best in the Stanford comparison. It is not a treatment for an anxiety disorder and we do not present it as one. It is useful for getting through the next ten minutes, which is a real and limited thing.

What else are you wrong about? Almost certainly several things. The crisis-detection failure we found in our own app during testing is documented in the report, including the part where our detector missed both phrasings of a crisis message. When we find the next one we will write it up the same way.

---

ILTY leads with a conversation and keeps the tools behind it, which is the whole argument in one sentence. Mr. Relentless is still the default and still the reason people stay.

Download ILTY and see what it offers before it offers you a breathing exercise.