This is the second of two essays drawn from a conversation I did with Kedar Nimkar. The first one covered some bits of rapid-fire round questions. You can watch the full episode below.
There were three threads from this conversation that I feel are worth writing about about. We spoke about premium repeatedly, the design curriculum at colleges problem that I've been wanting to write about for two years, and the smartness argument, which I hold a fairly strong view on that I didn't fully get to state. This post is on those three threads.
Designing for aspiration vs. price
In a Dharavi home without clean running water, without an adequate kitchen, the first large purchase when money becomes available is often a television. A big one large enough that it seems disproportionate to the room.
The television in that home (and in most homes) is doing work that nothing else in the space can do. It is social glue. It makes a small room a complete home. It is the thing your guests spot when they arrive. For a household where the living room is also the bedroom is also the kitchen, the television is the single object that says: this is a home, not just a space we occupy to sleep.
That is a rational premium purchase. The irrationality is in assuming that premium means the same thing across every household.
See, there's an India India that reads premium through size.
I grew up going to Chandni Chowk in Calcutta to get configured PCs. For years, monitor prices went up with screen size — a 17-inch cost more than a 15-inch, always. Then came the point when prices started climbing as screens got smaller with higher resolution. A 13-inch MacBook at twice the price of a 17-inch budget laptop. Two different Indias, operating simultaneously in the same market, reading premium through completely different lenses.
For the first India, size signalled investment. The bada hai toh behtar hai is a coherent response to a specific social context where what you own is seen by other people, and size communicates something that specs on a box cannot. The living room gets the renovation budget because the living room is where other people sit. The bedroom, on the other hand, is private. A product team that doesn't understand this will keep wondering why their beautifully minimal, better-specced, more expensive ₹12,000 television keeps losing to the bigger, cheaper ₹8,000 one when the answer is sitting right there on the wall.
There's a segment of the market where the number of visible camera lenses on the back of a phone is a legitimate signal of value. The ₹7,000 Android with four lenses on the back exists because it looked more premium than the three-lens one at the counter. This is correct design for that category. The purchase cycle is fast, the shelf life is short, and the buyer is deciding in minutes.
The error is applying this logic to a ₹18,000 RO water purifier. The person standing at that counter is not impressed by multiple LEDs indicating different features. They are worried about whether the filter will last, whether the service centre is reachable, whether this appliance will cost them more money in two years than it cost today. In its truest sense, multiple LEDs say: we did not know who we were building this for. Neither, for that matter, does the AI wash mode, the steam refresh cycle, or the seventeen-position dial on the machine that will sit in the utility area for the next decade.
Convenience is another dimension of premium in India.
In the interview, Kedar was wearing AirPods Max. There are better-engineered headphones at higher price points with bigger drivers, better leather, more ergonomic profiles, better cases. He wasn't using any of them. The reason could be the fact that AirPods connect to an iPhone without hassle. They switch between devices seamlessly. They pause when you take them off. None of these are audio features. All of them are why the product gets used every day. The team that built the H1 chip for seamless device switching was solving a problem that starts outside the headphone.
In the Indian home context, convenience premium looks like a mixer grinder knob you can turn with oil-coated hands because the grip was designed for that specific moment. These things will not appear in spec comparisons. All of them are the difference between a product someone trusts and a product someone tolerates.
And there's an India that reads premium through durability. At general stores in tier-2 and tier-3 cities, the first two questions are always: what is the warranty, and where is the nearest service centre. Because travel to a service centre costs money, travel means a missed day of work.
I have three drill kits at home - a Bosch, a Hoto, and a Xiaomi. The Hoto is objectively more beautiful. Satin TPU finish with great proportions, looks aesthetic as f*ck and is genuinely lovely. I'm always careful with it because I don't want to scratch the body. The Bosch one has scratches. I don't think about it when I pick it up. If you asked me to recommend one, I'd say the Bosch without hesitating. The Hoto is premium on the day of unboxing. The Bosch is premium because it's durable and with that definition, it'll always stay durable.
Okay, there's one more. The fascination of global CMF and Indian kitchens.
The global premium language of the last twenty years converged on white finishes, exposed brushed metal, minimal surface detail. Take Miele or Braun. Every European kitchen you've seen in a design magazine is calming but serious.
Now put those in an Indian kitchen. It'd look premium for about six weeks in an Indian kitchen. Then the mustard oil and turmeric get to it, the chimney finishes doing its sixty percent, and the white plastic becomes the colour of mild regret. What was premium on the day of unboxing has now become evidence that the design team had never seen an Indian kitchen in operation. However, matte dark finishes age well. In a product that stays in Indian homes for eight to ten years, aging well is aesthetics.
Industrial design schools aren't teaching right
When I hired the first couple of industrial designers onto my team, I asked them why they didn't know certain things. Then I looked at their curriculum. It's not good. I was being generous earlier on the podcast.
User surveys are not research.
I want to say this plainly because a lot of teams still run them, present the results in a deck, and then ship the product they were going to ship anyway. If you ask a customer to list five things they'd want improved, you will get five things they thought of in the moment, none of which are the things actually making their life harder, because those have been absorbed into muscle memory and no longer register as problems. They feel like the way things are.
The user persona is the same problem in a different font. The persona has a name like Sunita, 34, working mother, values convenience. Wake up! Sunita does not exist. The real person whose convenience you're supposed to be valuing has oil on her hands and something burning on the stove and did not agree to be a slide in your presentation. The survey and the persona together give you the feeling of having done research without the inconvenience of actually being in the room.
The only research that works is actually being in the room. Watching what the hands do. Watching where the body struggles and adjusts. The slapping of the water purifier lever four times before a bottle fills does not appear in any survey. It doesn't appear because the person filling the bottle has stopped noticing it. It appears only when you watch her, quietly.
Design schools are producing people who can render and not much else.
The sketch, the Keyshot render, the beautiful Behance case studies get the attention, sure. But, most of them have never stood in a factory production line studying the details. Most curricula have a unit on design thinking, a unit on user personas, a unit on sustainability principles, and a Figma (or equivalent) license. The factory, if at all, is an elective.
Manufacturing is treated as downstream some engineer's problem. This is how you produce designers who draw radii that are impossible to tool, wall thicknesses that warp on cooling, undercuts that add ₹40 per unit in tooling cost because no one asked whether the part could be drawn in a different direction. These decisions, made in ignorance at the sketch stage, travel all the way down the supply chain until they arrive in a customer's home as a quality problem. By which point the designer is presenting their next project at a graduate showcase.
Teardowns are the other half of this education which should be mandated. Take products apart, understand how they're constructed, read a product's entire set of priorities from the sequence in which it disassembles. This reading ability should be foundational for anyone in design schools. It comes only from doing it with real products and not from case studies.
Lack of deep context.
Take the Tupperware lunchbox snap. When you close the lid, the thud is solid and slightly louder than you expect. That snap is doing more work than the entire surface finish. It communicates sealed and well-made. Remove it from the same box and the box feels cheap, regardless of colour or texture or how much the CMF process cost.
We are conditioned to think friction is bad. Loud is bad. Resistance is bad. I find my the protein shaker from Boldfit that clicks shut more trustworthy than the one that just sort of closes. The drill case that snaps firmly is more premium than the one whose lid softly thuds. The microwave that beeps loudly when it finishes is doing useful work because it reaches you in the living room and tells you something happened. The refrigerator that makes the same sound is annoying because you are not waiting for the refrigerator. Same signal, two different contexts, two different correct answers. Context is the design problem. "Loud bad" is what happens when you stop thinking and start following rules.
My take on smart appliances
In the rapid-fire round, the question was which button I'd delete from most appliances. I gave two answers: the mode button, and the AI mode button.
The mode button is the hamburger menu of hardware products. It means the team couldn't finish the job of deciding what the product should do, so they handed a rotating list of options to the user and called it functionality. The AI mode button is the mode button's more dishonest successor. It appeared across washing machines, ACs, and air purifiers in the last two years. The marketing team knows it sounds premium and it does sound premium, right up until someone asks what it does and the answer is some version of "it optimises." Optimises what? For whom? The user presses it once, notices nothing different, and either leaves it on permanently or turns it off.
In the previous post, I gave the ceiling fan example. Here's another one. The refrigerator food-monitoring camera is the same logic taken to more expensive lengths. The pitch is your refrigerator will tell you when your food is going bad. But in most cases you know you put the mango there on Tuesday. It's been in there for over a week and you know exactly where the mango is in its journey (to a second death). This is what happens when the technology precedes the problem: you end up solving something that wasn't broken to begin with and adding three new things that can be.
Convenience backward, not smartness forward.
The frame I would replace the current definition with is simpler. Don't start with the technology you can add. Start with the friction you can remove.
Imagine you're at your desk and the lamp switch is across the room. You clap twice and the lamp comes on. This is technically interested but it found a real inconvenience and removed it. The technology is downstream of the problem. The problem came first.
The refrigerator camera started with the camera, not with a complaint about food management. The ceiling fan app started with the ability to build an app, not with a single user saying the regulator was the problem. The AirPods's chip started with the daily annoyance of switching between devices.
The lamp should work first.
A smart lamp that doesn't work reliably is a broken lamp with an app. The clap-activation doesn't matter if the light doesn't come on. Similarly, the food monitoring in refrigerator doesn't matter if the compressor fails at year three. The connectivity doesn't matter if the basic machine isn't worth connecting to. More components means more failure surfaces.
Every sensor, every wireless module, every PCB added in the name of smartness is a component that sits in the tolerance stack of everything around it and can fail at 2 AM in year six when the service centre is one hour away. The trade-off has to be justified by a real problem being solved, not by the desire to have something to put to impress self.
That is the only definition of smart that survives a real home.