Selling Products on Instagram Without a Website

Short answer: You can sell on Instagram with no website at all. Orders come through DMs or WhatsApp, payment through UPI, delivery through a courier aggregator with home pickup. A website becomes necessary when order volume makes manual tracking start producing mistakes, not before.

---

My first sale was Rs 340. A set of three handmade candles, going to a buyer in Nagpur.

Her entire message was: "is this available?"

It took me forty minutes to reply, because at that point I believed a DM reply had to sound professional. I wrote a paragraph about the wax, the burn time, the packaging, the dispatch window. She read it and said "ok send it."

The paragraph was wasted. She had already decided before she typed.

That took me an embarrassing number of months to understand. On Instagram people make up their mind from the feed, then open the DM. The DM is not where you convince someone. It is where you stop getting in their way.

The advantage of having no website

I used to think the missing website made me look like an amateur.

The opposite turned out to be true, at least early. Two friends of mine spent about six weeks building storefronts. Picking themes, getting payment gateway verification, resizing product photos to whatever dimensions the theme demanded. In those six weeks neither of them sold anything.

I spent that same stretch answering DMs. And answering DMs gave me something a website never would have: I could hear the actual question.

Somewhere around order forty I noticed that more than half of my buyers had asked exactly one thing, and it was not about the product. It was "how many days will it take." So I put the answer in my bio. Dispatch in two days, delivery in four to six. The question mostly stopped and orders went up a little.

A website would have hidden that from me completely. Confused visitors do not fill out a form explaining their confusion. They just leave.

The money mess

Everything went to my personal UPI at first. I would send my number, ask for a screenshot, done.

Two months in I opened my statement and found over 180 transactions I could not separate. Which one was an order and which one was my brother paying me back? A customer asked for a refund and I spent almost half an hour locating her payment.

So: separate account. And then, embarrassingly, a paper diary where every night I wrote the name, the city, the amount, and the dispatch date.

A paper diary. In 2026.

Honestly it beat every app I tried at that stage, because opening a spreadsheet on a phone at 11pm felt like work and a notebook did not. I use a proper sheet now. But those four scribbled months taught me that you should build the system when the pain shows up, not before, because building systems is a very comfortable way to avoid selling anything.

What actually gets seen

My best performing post was an ugly video. No editing, no music. I had spilled wax across the table and was scraping it off while the phone happened to be recording. Thirty-four seconds.

Eleven orders came from it.

Before that I had spent about three weeks on proper photography. I bought a ring light for Rs 1,190. White backdrop, clean shadows, the whole thing. Those posts brought in two orders total.

I am not saying good photos are useless. You need them the moment someone is deciding whether to pay. But the thing that stops a thumb is not polish, it is watching something happen. Hands in the frame. A mistake being fixed.

My rough pattern now is three process clips to one clean product shot. That ratio came from reading my own numbers, not from anyone's advice, and I suspect it would be wrong for a different product.

One more thing that surprised me. City hashtags did more than any general ones. Around a third of my early buyers were from my own city, and a few said outright that it felt safer buying local because there would be a real person to chase if something went wrong.

Mistakes that cost real money

I once accepted fourteen orders when I had stock for nine, assuming I would just make the rest over the weekend. I did not. Five people got their parcels nine days late, two asked for refunds, and one posted a screenshot of our chat on her story.

That day was genuinely awful. I refunded both and sent them an extra candle each. One of them ordered again months later. The other never came back, and to be fair, why would she.

Second mistake: showing shipping separately. For three weeks the product was Rs 299 with Rs 80 shipping mentioned in the DM. People kept evaporating right at that message. Then I made it Rs 379 with free delivery. Identical money. Orders roughly doubled.

I still do not fully understand what that does to people's heads. It works anyway.

Third one, and this is the one I would warn anyone about: for about six weeks I tried to answer every DM within five minutes. Phone in hand constantly, replying at one in the morning. My mood that month got bad enough that I seriously considered shutting the whole thing down. Now I reply in two windows, late morning and evening. Sales did not change. Not even slightly.

When the website becomes necessary

Fourteen months in, I still have not hit that point. But I can see it from here.

The signal will be volume. Once it is consistently more than a dozen orders a day, or once a repeat buyer asks for the same thing she ordered last winter and I have to go digging through a sheet to find out what that was, the manual setup stops being clever and starts being negligent.

For now it holds. Instagram, WhatsApp, one spreadsheet, one courier app. Last month was 47 orders. The best month so far was 61, in December, and I nearly broke myself getting those out.

Yesterday someone messaged from Kochi asking whether I had a website because she wanted to browse properly. I said no and sent the catalogue as images. She ordered anyway.

Which was nice. But I keep thinking about how many people ask that question silently and just close the app, and there is no way for me to ever know that number…