HomeOur story
We built the thing we kept failing to build with willpower
Spark is a separate physical object because the only method that ever reliably worked for us was leaving the phone in another room — and no app can put itself in another room.
The prototype was a kitchen drawer
Long before there was hardware, the method was embarrassingly low-tech: put the phone in a drawer in another room and shut it. It worked far better than any app we had tried, which was annoying, because we build software for a living.
The problem was that it was all-or-nothing. A phone in a drawer is also a phone that cannot take a call from school, show a boarding pass, or pay for coffee. We kept reaching for it for a legitimate reason and then losing forty minutes to something else entirely.
The insight was about direction, not force
What the drawer got right was not that it was strict. It was that it was asymmetric: putting the phone away cost nothing, and getting it back cost a walk. Every piece of software we had tried made both directions cost the same — one tap in, one tap out — and so the tired version of us always won.
We did not need a stronger lock. We needed the key to be somewhere else.
We are not the only people to have wondered about this. One well-known experiment found that people who left their phone in a different room did better on working-memory tasks than people whose phone sat face-down and switched off on the desk — though a later pre-registered replication did not find the same effect. We ran neither study, and we have run none of our own; it is simply the closest description we have found of what the drawer seemed to be doing. The studies, both sides, all of them other people’s.
So we separated the two
Spark doesn’t move your phone; it moves the unlock. You choose what gets blocked, and calls always get through. Locking stays instant and free — one tap on Lock now, or a schedule — and it never needs the key, because hesitating to lock is the real failure mode. Unlocking is the part that needs the small key you deliberately left in the other room: you walk to it and hold your phone against it. The app is free; the key is a one-time purchase and there is no subscription behind it. What that costs and what arrives.
How it started
Puneet and Akshay started building Spark in 2026, after noticing the same thing kept happening to both of them: a two-minute check of the phone mid-task that quietly ate the rest of the hour. Neither of them trusted willpower to fix it, so instead of another app with another off switch, they built a key you have to get up and walk to.
The tag became a key
Spark did not begin as a Bluetooth device. The first version was an NFC tag: you tapped it to the phone and the apps unlocked. It was pleasingly physical, and it did not work on a large number of phones — many budget Android handsets ship with no NFC reader, and India, our first market, has a great many of them. So we built a Bluetooth key too. Today there are two: the Bluetooth Spark, for any phone on iOS 17 or Android 12 and later, and the NFC Spark, slightly cheaper, for phones that have a reader.
Neither is a vault. The NFC Spark is a standard tag, and a standard tag can be copied; the Bluetooth Spark’s broadcast can be recorded and replayed by someone with the right kit. We say so because it doesn’t change the job: Spark is built to beat a moment of weakness, not an attacker.
What did not change is the gesture. You unlock the Bluetooth Spark the way you unlock the tag — phone held flat against it, for about fifteen seconds until the ring in the app fills — so radio range never comes into it; the distance that matters is the walk to wherever you left it. The one new cost was a battery, a CR2032 coin cell, and a flat battery must never mean a bricked phone. That is part of why emergency unlocks exist, and why there are five for life rather than an endless supply — enough to survive a dead key, few enough that they cannot quietly become the normal way you unlock.
The engineering write-up of the move from NFC to Bluetooth — what we lost, what changed, and the battery question.
What we refuse to do
A focus product built on the same manipulative patterns as the apps it blocks would be a bad joke. That is a design constraint here, not a marketing line:
- No streak mechanics designed to make you anxious — no notifications nag you to protect a streak.
- No selling your usage data. Blocking runs on your own phone, through the operating system’s official APIs.
- No daily notification asking how your screen time is going.
- No plan in which paying us more makes the blocking work better.
- No claiming enforcement we cannot deliver. Android permits stricter enforcement than iOS does, and where a platform limits what we can block, we write that down plainly rather than burying it in a footnote. What each platform actually allows.
Who we are
The registered entity, address and GSTIN are published in full on the policy pages, because those are binding. The people behind it:
Puneet builds the product and most of the code behind it, and started Spark after one too many focused mornings lost to a phone he only meant to check for thirty seconds.
Akshay runs the business side of Spark — sourcing, support and the parts that keep the shop honest — and is usually the person who replies when you email us.
Put the key in another room.
Locking is one tap in the app and never needs the key. Unlocking means walking to the key and holding your phone against it — which is the whole idea, and the only part we could not build in software. The app is free; the key is the one-time buy.