About Varna Code
Why I Built Varna Code
I grew up in Jaipur, in a city that has been dyeing cloth for centuries.
Bandhani, leheriya, block prints. Bazaars stacked with indigo and madder and marigold.
And I still had no idea which colours suited me.
I’m a dark-skinned guy. For most of my life, shopping went the same way. I’d find something in a colour I liked, hold it against myself in a badly lit trial room, and have no real sense of whether it worked. So I’d put it back and buy another black t-shirt. Or grey. Or whatever I’d bought last time.
It wasn’t that I didn’t care. It’s that I never knew.
I’m also, honestly, a bit shy about it. Experimenting with clothes means walking out of the house in something you’re not sure about and finding out in public whether you got it right. Some people enjoy that. I don’t. Every wrong buy felt like money gone and one more small thing to be self-conscious about, so eventually I stopped trying. Black and grey are safe. Safe is easy to hide in.
And I didn’t want it to become a project. I wasn’t going to spend weekends scrolling through a thousand products, or hours in the bazaar, or a lot of money working out what suits me by trial and error. I wanted someone to tell me: these are your colours, start here.
So I went looking for that.
I couldn’t find it.
There’s no shortage of colour advice online. But nearly all of it rests on one of two systems — the Western “seasonal” method, developed around fair, Caucasian skin, or the Korean personal-colour system, which is hugely popular now and calibrated for East Asian skin.
Both are genuinely good systems. Neither was built for me.
Search for advice for Indian skin and you mostly get “wheatish” — a word that describes almost everyone and helps almost no one — along with a lot of quiet suggestions about what not to wear. Search for anything deeper than that and the results thin out fast. There was no tool, no app, no site that would take a skin tone like mine and give me a straight answer.
That gap is the entire reason Varna Code exists.
So, I built the thing I’d been looking for.
Varna is Sanskrit for colour. That’s the whole idea. Not a category, not a ranking, not a euphemism. Just colour, and which ones work with yours.
Varna Code sorts skin into eight tones — Ivory through to Cacao — and tells you which colours genuinely lift your complexion, in your size, and where to buy them. It takes about a minute. You don’t need to know any theory. You don’t need to feel confident about clothes.
I should be straight with you about something. I’m 31, and I still find buying clothes hard. This isn’t a story about a problem I solved and am now selling you the answer to. It’s a problem I still have. The difference is that I no longer start from zero — I know my tone, I know my eight colours, and I know which ones to walk past. That turned out to be enough to make it bearable.
I’m not a stylist. I’m not a fashion expert. I’m someone who got tired of guessing, decided to work it out properly, and then built it into something anyone could use. I use it myself. That’s the honest test.
There’s something I still find strange about all this. I come from a place that understands colour better than almost anywhere — generations of people who know exactly what a dye does in sunlight, what red means in October, how to read a fabric by touch. All of that knowledge existed around me my whole life. None of it was ever pointed at the question of what suits me.
That’s what I’m trying to fix.
If you’ve ever stood in front of a mirror holding up a shirt with no idea whether it works — this was built for you. Every shade is beautiful. Every shade is different. You just need to know which one is yours.
— Shashank, Jaipur · founder, Varna Code
What Varna Code Gives You
- A free skin tone and colour matching quiz — about a minute, no sign-up
- Your personal palette: eight colours that work with your tone, and the ones to walk past
- A wardrobe plan built around your Tone and occasion
- Real shopping links to Indian retailers, at prices you’ve set
- Colour guidance based on actual colour science, not guesswork
Where This Is Going
I’m starting in India, because that’s where I am and that’s the gap I know best. But skin like mine isn’t only in India — it’s across South Asia, the Gulf, Southeast Asia, and in every diaspora community that grew up being told to avoid bright colours. Varna Code is being built for all of them, eventually.
Say Hello
Found a colour combination that works brilliantly for you? Think I’ve got something wrong? I’d genuinely like to know — write to me at info.varnacode@gmail.com.

That’s me, in the kind of ordinary afternoon light this whole site is about.
It felt only fair to run my own system on myself. Sampling my cheeks gives an ITA° of about 22, which puts me in the Tan band — Caramel, with a warm undertone. So of everything on this site, the colours I should be wearing are the deep olives, the burnt oranges and the warm golds.
I’ll be honest about the limits, because they matter more than the result. My first measurement came back Espresso — it had accidentally sampled my hair instead of my skin. And my right cheek reads warmer than my left, because the sun was on that side. A phone photo, with automatic white balance and nothing to correct against, can move the answer by a whole band. And the reading has a season. That ITA° of 22 was taken in winter. By the end of an Indian summer I sit closer to Espresso, two bands deeper — which swaps five of my eight colours and reverses my neutrals entirely: the chocolate that works against my face in December is on my avoid list in June. So the honest answer is that I am Caramel in winter and Espresso in summer. Any single reading of me is a reading of one day.
That is exactly why this site does not ask you for a selfie. A photograph can be read wrong in a dozen ways — the light, the camera, the white balance, which side of your face the sun was on. I would rather be honest about what can and cannot be measured from one.
>
The colour data is public
Every number this site runs on is published openly. Eight skin tones, twenty-four palettes, 192 colour values with hex codes, the seasonal neutrals and the pairing rules — all on GitHub as an open dataset under CC BY 4.0, free to use with attribution.
It also carries a methodology file that lists what is weak about it, including a tone whose palette overlaps its neighbour too closely and a set of neutral values that were specified rather than measured. If someone can improve on it, I would rather they did.