The claim it is not: a networked device. SOSA publishes no Wi-Fi, no Alexa, no Google Assistant, no Matter, no Zigbee, no Z-Wave, no KNX, no Crestron, no Control4, no RS-232, no dry contact and no API. It automates itself — Bluetooth app, onboard buttons, 1h / 4h / 8h / 24h windows, day selection, adjustable intensity, auto-stop, key-lock — and it runs alongside the automation, not inside it.
Why it has stayed outside: no standard, therefore no protocol; a consumable that has to be physically refilled; and a broken feedback loop — there is no cheap reliable ambient scent sensor, and the human sensor stops reporting within minutes of exposure.
What would have to change: a measurement standard, a device profile in a mainstream control model, and a level-reporting answer for the refill. SOSA publishes none of these and I am not going to imply a roadmap.
2. No, as a network layer — and pretending otherwise is how specifiers get burned. Vaayu automates itself: a Bluetooth app and onboard buttons, run windows of 1h / 4h / 8h / 24h, day selection, adjustable mist intensity, auto-stop and a key-lock, all stored in the unit. SOSA publishes no Wi-Fi, Alexa, Google Assistant, Matter, Zigbee, Z-Wave, KNX, Crestron, Control4, RS-232, dry contact or API. It cannot join a scene, be triggered by a sensor, be voice-controlled or report its state, and it never will unless something in the wider category changes.
3. Four things kept it outside, and only one of them is about ambition. There is no interoperability standard for scenting, because the category never generated the commercial pressure that produced one for lighting. There is therefore no protocol. There is a consumable — a 400ml tank emptying over about 90 or more days — and no control system can pour oil. And there is no working feedback loop, which is the deepest reason and the one nobody talks about.
4. The broken feedback loop is the real barrier. Lighting closes its loop on a lux sensor, climate on a thermostat, audio on a level meter. Scent has no affordable, reliable ambient sensor with an agreed unit of measurement — and the one instrument that does exist, the human nose, stops reporting a constant stimulus within minutes. A closed-loop scenting system would be steering by an instrument that goes blind on contact with the thing it measures.
5. Which turns out to matter less than it sounds, because the desirable behaviour is constancy. In lighting we prize responsiveness. In scent the luxury signal is a low, unchanging level that a person notices on arrival and then stops noticing. A fragrance that visibly changes state is one somebody has detected as a mechanism. So the schedule is not a reduced form of integration for this subsystem; for this application it is very nearly the whole of what integration would have been used for.
Vaayu is ₹11,999 per zone, covers up to 1000 m³ of connected volume, runs under 38 dB, draws 5W at DC 12V / 1A, holds a 400ml sealed residue-free tank, and is CE, RoHS and SGS certified. Free shipping above ₹499, and a portion of every order supports girl-child education through Nanhi Kali.
The case for it being a real layer, and not an accessory
I want to make the argument properly rather than assert it, because this reader has heard a great many categories described as the next thing. Three reasons hold up under pressure, and one of them is commercial rather than sensory.
Vaayu₹11,999A luxury residence at this level has already had every visible and audible decision taken deliberately: colour temperature, beam angle, acoustic treatment, materials, temperature setpoints, the sound a switch makes. Scent is the only channel left where nothing has been decided, which means it is also the only one where a decision produces a change nobody expected. That is unusual value at this stage of a project. It is also why the phrase "the house smells expensive" is so hard to argue with and so easy to achieve badly — the same composition at too high a concentration flattens into one sweet impression, which is what people mean when they say air freshener. Restraint is the luxury signal here, and it costs nothing.
Cold-air oils₹1,799Every other layer announces itself. A lighting scene is visible, a temperature change is felt within minutes, audio is unmistakably present. Scent works at the level a person cannot narrate: the nose reports change far better than steady state, so a scented building is registered in the two seconds of crossing a threshold and then disappears from conscious awareness entirely. That is the design target, not a failure — a fragrance somebody can smell continuously all evening is set too strong for the volume. The correct dose is one that a person entering notices and a person sitting forgets, which means dosing for the arriving nose rather than the occupying one, and it is the single most common mistake in this category to do the opposite.
Ask before specifyingContactAt ₹11,999 a zone, plus a 400ml fill running up to about 90 or more days and cold-air oil at ₹1,799 for 300ml, this sits low against every other line on a premium automation package, and it is the line a client is most likely to mention to a guest. It is also worth being clear about what the money buys, because it answers the question people are too polite to ask: you are paying for transport, control and evenness rather than for a better smell. A modest reed diffuser and an eleven-thousand-rupee nebuliser can carry closely related compositions and produce completely different buildings, and the difference is entirely in how the material is moved and how reliably it arrives. On the commercial arrangements themselves — there is nothing published: no dealer programme, no margin, no trade price, no territory, no minimum order, no lead time, no installation charge and no maintenance contract. Ask through the contact page rather than assuming terms.
The four reasons scent stayed outside the automation stack
Now the part that makes the argument credible, because a category that claims to be next has to explain why it was not already. Four reasons, and they are structural rather than accidental. There is no interoperability standard for scenting. Lighting, shading and climate converged on shared control models because thousands of manufacturers needed each other’s products to work together and the market punished those that refused. Scenting never generated that pressure — it is a small category, most machines are sold as standalone appliances, and no purchaser was ever blocked from buying a diffuser because it lacked a device profile. With no standard there is no protocol, so every machine keeps its own logic and its own app, which is exactly the position this product is in and I would rather describe it than dress it up.
The third reason is the consumable, and it is the one most easily underestimated by people who work with luminaires and thermostats. Nothing else on the bus runs out. A fragrance machine holds 400ml and empties over about 90 or more days depending on intensity and scheduled hours, which means the genuine maintenance event on a scenting installation is a person walking into the building with a bottle. No protocol solves that. An integration might tell you the tank was low — more on that below — but somebody still has to arrive, open the unit, refill it and record the date, which is why refill access is a design decision and why a synchronised quarterly cadence across all units on a property beats an accurate one.
The fourth is the one I find genuinely interesting, and it is the reason I do not expect this to resolve quickly. The feedback loop is broken at the sensor. Every other layer of a smart home can be measured while it is running: a lux sensor reads light, a thermostat reads temperature, a meter reads sound pressure, and the control system corrects toward a target. Scent has no affordable, reliable ambient sensor and no agreed unit in which to express a target. The only instrument that matters is a human being, and a human being adapts to a constant stimulus within minutes — the signal still arrives, the brain stops reporting it. So a closed-loop scenting system would be steering by an instrument that goes blind on contact with the thing it is measuring, and would be doing so in service of a variable that is not supposed to change anyway. That is not a gap waiting for a firmware update. It is a property of the sense.
The barriers, and what each one actually blocks
Read as an engineering problem rather than a complaint, because that is how this audience will assess it. The right-hand column is deliberately conservative: SOSA publishes no roadmap, and nothing in this table should be read as a commitment by anybody to build any of it.
| Barrier | Why it exists | What it blocks | What would have to change | Where it stands |
|---|---|---|---|---|
| No standard | A small category with no interoperability pressure | The device does not exist in any product library | An agreed device profile in a mainstream control model | Nothing published |
| No protocol | A consequence of the first — every machine keeps its own logic | Scene membership, triggers, status, remote editing | A standard first; a protocol follows it, not the reverse | SOSA publishes none of any kind |
| A consumable | Oil is physically used up — 400ml over about 90+ days | Unattended operation beyond a quarter; true set-and-forget | Level reporting at minimum; the refill visit remains regardless | No tank telemetry exists in this product |
| A broken feedback loop | No affordable ambient sensor; the human nose adapts within minutes | Any closed-loop or self-correcting behaviour | A reliable sensor and an agreed unit of measurement | A property of the sense, not a firmware gap |
| The one that is not a barrier | Dynamic control — the thing automation is best at | Nothing worth having: constancy is the design target | Nothing. A scent that changes state is one somebody has noticed | Already correct as it is |
Vaayu · per zone₹11,999See specification → 
Aangan · larger volumes₹25,999See it → Cold-air oils · 300ml₹1,799Shop →
What would have to change, stated without a roadmap
If this category is going to come inside the automation stack rather than sit alongside it, three things have to happen and they have to happen in this order. First, a measurement standard. Nothing can be automated in a closed loop until there is an agreed way to say how much of it is present, and scent has no such unit in ordinary use and no affordable sensor to produce one. Until that exists, a scenting device on a bus can be commanded but never verified, which is a considerably weaker proposition than it sounds — you would gain the ability to switch fragrance from a panel while keeping no way at all of knowing what the room actually smells like.
Second, a device profile inside a mainstream control model, so that a scenting machine is an object an integrator can drag into a project the way a dimmer or a shade motor is. That is a standards-body activity rather than a manufacturer activity, and it follows commercial pressure rather than leading it. Third, an answer to the consumable: at minimum a reliable level report, so that a facilities manager can see remaining oil across six units in a large villa without walking to each one. I would genuinely like that third one to exist — it is the limitation I feel most on multi-unit properties — and it does not exist in this product. SOSA publishes no roadmap, no forthcoming integration and no planned protocol support, and I am not going to imply one on a page an integrator might rely on. What we publish is what the machine does today.
Meanwhile there is one thing that already works in exactly the way people imagine integration would, and it deserves the last word in this section because it is chronically underrated. Mounting into an HVAC return is a real, published, mechanical integration: the building’s own moving air becomes the distribution system, which is why a well-executed installation has no findable source at all — no object on a table to walk towards, no gradient to climb. That absence of a locatable source is precisely what people are describing when they say a building "does this by itself". It is achieved with ductwork and a schedule rather than with a protocol, and for the overwhelming majority of clients it is indistinguishable from the thing they were imagining.
What to do in the meantime — and the honest gap
The practical position for an integrator or an architect who accepts the argument, in the order it is used on a project. The final row is the disclosure that has to be made in writing, by the person who carries the client relationship, at the beginning rather than at commissioning.
| Move | What it means in practice | Why it is the right move now | Reference |
|---|---|---|---|
| 1. Specify it as a scheduled subsystem | An independent appliance per zone, commissioned once and key-locked | It is what the product is, and it behaves like a correctly commissioned subsystem that never needs touching | Vaayu ₹11,999 per zone |
| 2. Reserve at drawing stage | A 5W low-voltage point at the mounting position and a serviceable location | Costs almost nothing and keeps every option open, even for an undecided client | DC 12V / 1A |
| 3. Zone in volume | Connected volumes with doors in their normal position, ceiling heights annotated | A closed door is a boundary; an open plan is one volume | Up to 1000 m³ per unit |
| 4. Hold one accord | One composition across the property, varied by intensity rather than by scent | Recognition across visits, and no corridor where two ideas meet | Cold-air oils, 300ml ₹1,799 |
| 5. Dose down | Lowest useful intensity, judged from the doorway, then locked | Almost every complaint in this category is a dosing complaint, and almost every dosing error is upward | Key-lock, auto-stop |
| 6. Diary the refill | A synchronised quarterly visit across all units, with settings recorded | A 400ml fill runs up to about 90+ days depending on intensity and hours; nothing will remind you | 400ml sealed tank |
| No protocol, no roadmap: the honest gap | SOSA publishes no Wi-Fi, Alexa, Google Assistant, Matter, Zigbee, Z-Wave, KNX, Crestron, Control4, RS-232, dry contact or API, no tank telemetry, no cross-unit dashboard and no forthcoming integration of any kind | Put it in the client’s file in writing at the start. They will accept the limitation; they will not forgive discovering it at handover | Ask us first |
Versailles
I am asked this question in two very different tones. From an integrator it is usually sceptical, and fairly so: they have watched several categories announce themselves as the next layer and then fail to produce a device profile. From a client it is usually enthusiastic and slightly vague. My answer is the same to both, and it is deliberately split in half: yes to the experience, no to the network, and anybody who blurs those two is going to have an uncomfortable conversation at commissioning.
What convinced me it is a real layer rather than an accessory was watching how people describe buildings they cannot analyse. Nobody says a room smelled correct. They say it felt considered, or expensive, or calm, and a surprising amount of the time what they are reporting is something they smelled for two seconds while walking in and never consciously registered again. That is a design channel, and at the moment it is the only one in a premium home that nobody has been given responsibility for.
What stops me overselling it is the sensor problem, which I do not expect to be solved soon and which I would rather explain than skirt. There is nothing cheap and reliable that measures scent in a room, and the only good instrument — a person — adapts within minutes of walking in. So there is nothing for a control system to correct toward, and there would be nothing worth correcting anyway, because in this discipline a level that never changes is the goal rather than the fallback. Everything is composed in Pune, and a portion of every order funds a girl’s classroom through Nanhi Kali.
Frequently asked questions
- Can fragrance be automated like lighting, curtains and air conditioning? — scheduled, not integrated.
- What an automated scenting system actually is — reservoir, nebuliser, schedule, intensity, mounting.
- Can a smart home turn its fragrance on and off? — the machine owns the logic, not the house.
- Is a waterless machine suitable for a smart home? — no humidity, no residue near electronics.
- How it fits a luxury setup — placement in the project rather than in the protocol stack.
- Scheduling it to certain times — the 1h / 4h / 8h / 24h windows through a real week.
- Different intensities in different areas — a machine per zone, each set independently.
- Planning it during design — the supply, the return and the zone boundaries.
- Lighting, climate and scenting together — a shared timetable, not a shared bus.
- Brand: the SOSA founder story.
SOSA Vaayu — facts verified August 2026: ₹11,999. Waterless cold-air nebulising diffuser; pressurised air shatters undiluted oil into a dry nano-mist, with no water, no dilution, no added humidity and no residue. Coverage up to 1000 m³, approximately 2,000–3,000 sq ft at standard ceiling height, read as connected volume rather than floor area. Control: Bluetooth app plus onboard buttons, run windows of 1h / 4h / 8h / 24h, day selection, adjustable mist intensity, auto-stop and key-lock; the schedule is stored in the unit. SOSA publishes no Wi-Fi, Alexa, Google Assistant, Matter, Zigbee, Z-Wave, KNX, Crestron, Control4, RS-232, dry-contact or API integration of any kind, no tank telemetry and no cross-unit dashboard; Vaayu cannot be started, stopped, scheduled, triggered, grouped or monitored by any home-automation control system, voice assistant, hub or sensor. Mounting: freestanding, wall-mounted or into an HVAC return — a mechanical integration, not a digital one. Power DC 12V / 1A, 5W at the mounting position. Under 38 dB. 400ml sealed, residue-free, refillable tank lasting up to about 90+ days depending on intensity and scheduled hours; SOSA publishes no millilitres-per-hour consumption rate and none is estimated here. CE, RoHS and SGS certified. Black and white finishes. Oils must be alcohol-free, water-free fragrance oils formulated for cold-air diffusion, such as the SOSA Hotel Collection (100ml ₹999 · 300ml ₹1,799 · pack of seven ₹1,799); thick undiluted essential oils are not recommended as they clog the fine nozzle. Other SOSA machines: Sukoon ₹1,899 ultrasonic (500ml, 270–320 sq ft), Boond ₹899 (up to about 150 sq ft), Megh ₹3,499 (six-litre tank, about 100 hours runtime, only about 215 sq ft — a runtime and humidity machine, never a coverage upgrade), Aangan ₹25,999 (~8,000–10,000 sq ft), Meenar ₹38,500 (12,000–18,000 sq ft). SOSA reed diffusers are alcohol-free, phthalate-free, paraben-free, IFRA-compliant and 0 ppm formaldehyde, with six fibre reeds in a refillable glass bottle, composed in Pune. Fragrance does not remove or neutralise an existing odour; the correct sequence is source, then ventilation, then fragrance. A scent machine adds fragrance; it does not clean, purify or deodorise air, and SOSA makes no health or air-quality claim. SOSA publishes no dealer programme, reseller margin, trade price, distributor agreement, territory, minimum order, lead time, installation charge or maintenance contract; commercial enquiries are welcome at https://sosahomeandbody.com/pages/contact. Free shipping above ₹499. A portion of every order supports girl-child education through Nanhi Kali. SOSA is an independent Indian fragrance house; the Hotel Collection scents are SOSA’s own interpretations inspired by the world’s finest hotels, and SOSA is not affiliated with or endorsed by any hotel brand or any home-automation manufacturer. Prices subject to change — see the live product pages.