There is a version of this guide that runs to forty steps. It includes choosing a colour palette and registering a trademark. It is also the reason most people never launch.
What follows is the short list. Nine decisions, in the order they bite. Get these right and the rest is admin you can do in an evening. Get them wrong and no amount of theme-tweaking saves you.
1. Decide what you are actually selling
Not the category. The unit. "Ceramics" is not a unit. "A 350 ml glazed stoneware mug, ₹850, ships in a 500 g box" is a unit, and it is the thing every later decision depends on.
Write down, for each product you are considering:
| Field | Example | Why it decides things later |
|---|---|---|
| Selling price | ₹850 | Sets whether free shipping is survivable |
| Cost to make or buy | ₹310 | Your real margin before shipping |
| Packed weight | 480 g | Drops you into a courier weight slab |
| Packed size | 12 × 12 × 12 cm | Volumetric weight can override actual weight |
| Fragile? | Yes | Adds ₹15–₹30 of packaging per order |
| Restock time | 5 days | Decides whether you can promise 2-day dispatch |
If you cannot fill that table, you are not ready to build a storefront. You are ready to make one more sample.
2. Decide where the store lives
There are four honest options in India, and they trade off in a predictable way.
| Option | Monthly cost | Set-up effort | Who it suits |
|---|---|---|---|
| Instagram + DMs only | ₹0 | An afternoon | Testing whether anyone wants it at all |
| Marketplace (Amazon, Flipkart, Meesho) | Commission 5–25% per order | A week of catalogue work | Products that compete on price and are already searched for |
| Hosted store builder | ₹1,500–₹7,000 | 2–5 days | Anyone who wants their own brand and repeat buyers |
| Self-hosted WooCommerce | ₹400–₹2,000 plus your time | 1–3 weeks | People who genuinely enjoy maintaining a site |
Most first-time sellers should not start on a marketplace. The commission is survivable; the problem is that you learn nothing about your own customers and cannot email them again. A marketplace is a distribution channel you add later, not a home.
If you want the full cost comparison with real numbers, we did one: Shopify vs WooCommerce vs BizRyn for Indian sellers.
3. Decide how people pay you
Three methods cover more or less everything in India, and they behave very differently.
Turn on UPI first, cards second, and think hard about cash on delivery. COD still drives a large share of orders outside metros, and sellers who refuse it often lose those orders entirely — but it also brings returns-to-origin, which is the single most expensive thing that can happen to a small store. We wrote up the RTO maths separately in cash on delivery without losing money.
4. Decide your shipping promise before you decide your shipping partner
Pick one sentence and put it on every product page:
- "Dispatched in 2 working days, delivered in 3–7."
- "Made to order, dispatched in 7 days."
- "Same-day dispatch on orders before 2 pm."
The promise sets everything else: whether you need stock on hand, which courier tier you buy, whether you can offer free shipping. Sellers who skip this end up promising nothing, and a page that promises nothing converts badly.
Then work out what it actually costs. Courier pricing in India is slab-based, not linear, and a 520 g parcel costs the same as a 1 kg one. That single fact has more effect on small-store margins than almost anything else, and it is why shipping rates decoded is the most re-read piece we have.
5. Decide whether you need GST today
The short version, and please confirm it against your own turnover with a CA:
The full breakdown, including HSN codes and what changed in the September 2025 rate reform, is in GST for online sellers.
6. Decide your packaging as a cost line, not an aesthetic
Packaging is where a good margin goes to die quietly. A branded rigid box at ₹48 on a ₹700 order is 7% of revenue, and nobody has ever left a review because the box had a logo on it.
Start with protection, not print
Bubble wrap, a right-sized corrugated box, and tape. If the item survives a 1 m drop in the box, you are done. Everything after this is marketing spend.
Right-size to the weight slab
A slightly smaller box that drops you from 750 g to 480 g saves more per order than any packaging discount you will negotiate in year one.
Add one cheap human touch
A handwritten card costs ₹2 and gets photographed more often than a ₹50 box. This is not a theory; look at what customers actually post.
Only then consider printed material
When you are shipping 300+ a month and the unit price of printing collapses, revisit it.
7. Decide what your product page has to say
Almost every first store under-writes its product pages. A visitor who cannot answer these seven questions in ten seconds leaves:
- What exactly is it, in one line?
- How big or how much is it? (In numbers. "Medium" is not a size.)
- What is it made of?
- What does it cost to get it to me, and when will it arrive?
- What happens if it breaks or I hate it?
- Has anyone else bought one?
- How do I ask a question before I buy?
The product description guide turns each of these into a template you can fill in. Photos matter at least as much, and you do not need a studio — a phone, a window and a white sheet outperform most rented setups if you use them properly.
8. Decide how the first fifty people find you
This is the step almost every guide skips, and it is the one that decides whether you have a business in month three.
Your first 20–50 orders will not come from search. They will come from:
- People who already follow you personally.
- WhatsApp groups where you are a member, not a marketer.
- One local reseller, market or café that agrees to carry three units.
- A ₹200–₹300/day Instagram test, once you have something worth testing.
Search traffic is real, and it compounds beautifully. It just arrives in month six, not week one. Plan the bridge. We mapped where early orders genuinely come from in your first 100 orders.
9. Decide what you are going to stop doing
This is the one nobody frames as a decision, and it is the reason stores stall at 30 orders a month.
By order forty, you will be spending most of your week on: confirming orders, answering the same nine questions, booking pickups, chasing one late parcel, reconciling what got paid out, and updating stock. None of it is the work you started the business to do, and all of it grows linearly with sales.
You have three options. Hire someone part-time at ₹8,000–₹15,000 a month. Do it yourself and cap your growth at whatever your evenings can absorb. Or hand the repeatable parts to software that does them with your rules attached.
First-week checklist
That last one matters more than it sounds. The list of things that annoyed you on day seven is the list of things to automate or delete by day thirty.
Frequently asked questions
How much money do I need to start an online store in India?
Realistically ₹8,000–₹25,000 for the first month if you are making or sourcing in small quantities. The software is the cheapest line on that list; inventory, packaging and a small ad test are the rest. The full budget breakdown has the line items.
Do I need a registered company to sell online?
Not to start. Many sellers begin as a sole proprietor with a current account in the business name. You will want a proper structure before you take on staff, raise money, or hit a turnover where the tax treatment changes — talk to a CA at that point, not before.
Should I sell on Instagram or build a website?
Both, in that order. Instagram tells you whether anyone wants the thing. A website is what stops you from manually replying to "price?" four hundred times and lets people buy at 11 pm without you.
How long before I get my first order?
If you tell people directly, days. If you launch quietly and wait for traffic, months. The variable is not the store quality; it is how many humans were told.
What is the most common mistake first-time sellers make?
Pricing for the product and forgetting the parcel. A ₹400 item with a ₹70 shipping slab and a 6% RTO rate is a loss-making product that looks profitable in a spreadsheet.