Right now your stock is probably in a spare room. Cartons against a wall, a few on the wardrobe, the fast movers near the door so you can grab them at 11pm when an order drops. It works. It costs you nothing but space you already pay rent on.
Then one of three things happens. You run out of room. Or you land a supplier order so big it won't fit in the flat. Or you go on Amazon and they tell you to ship stock to their warehouse. Suddenly the question isn't "how do I pack faster," it's "where should my inventory physically live?" That's a different decision, and it has real money attached to it.
This guide is only about that: where the stock sits. Not which courier to use, not how much to reorder, and not the separate "should I stop packing orders myself" question, which is the fulfilment call covered in choosing a 3PL for D2C in India. Just the four places your inventory can live as you grow, what each costs in rupees, and the volume where each stops making sense. Get this wrong and you either drown in dead space you're paying for, or throttle your growth because there's nowhere to put stock.
Your stock lives in one of four places, and you graduate through them by volume. Home or garage: free, until roughly 20 to 30 orders a day, when it eats your house and evenings. Rented self-warehouse: a godown lease (₹15 to ₹30 per sq ft a month in metro peripheries) plus hired hands; makes sense when you want control and have steady volume but not enough to justify a 3PL's margin. 3PL fulfilment centre: they store, pick and pack for you, charging storage plus per-order pick-pack; the default once you're past ~1,000 orders a month. Marketplace FCs (Amazon FBA, Flipkart, Meesho): stock sits inside the platform, cheap off-season but brutal in Oct-Dec peak (roughly 3x) with penalties on stock past 180 and 365 days. Most brands run a hybrid: a 3PL or self-warehouse for the website, FBA for the Amazon slice. Go multi-location only once volume and zonal data prove it. Never buy stock you can't store cheaply and sell inside 90 days.
The four stages of D2C warehousing in India
Almost every brand walks the same ladder, climbing rungs as volume forces you up. Here's the whole map in one table, then the honest detail on each.
| Stage | Where stock lives | Rough volume it fits | What you pay | Biggest risk |
|---|---|---|---|---|
| 1. Home / garage | Your flat, a spare room, a garage | Up to ~20-30 orders/day | ₹0 extra (space you already rent) | Runs out of room; eats your time and home |
| 2. Rented self-warehouse | A godown you lease and staff | ~30 to a few hundred orders/day | Rent (₹15-30/sq ft/mo) + staff + setup | Fixed cost regardless of sales; you run ops |
| 3. 3PL fulfilment centre | A partner's warehouse | ~1,000+ orders/month | Storage + per-order pick-pack + shipping | Less control; SLA and accuracy depend on them |
| 4. Marketplace FC (FBA etc.) | Inside Amazon / Flipkart / Meesho | Any, for that channel's orders | Storage (peak spikes) + fulfilment fee + penalties | Peak storage cost + aged-stock penalties |
One thing before the detail: these aren't either-or once you scale. The common end state is a hybrid, where website and D2C orders ship from a 3PL or your own godown while the stock you sell on Amazon sits inside FBA. More on that split below.
Stage 1: Home or garage (₹0, until it isn't)
This is where you should start, full stop. At 10 orders a day, storing stock at home costs nothing beyond space you already pay for, and it keeps you close to the product. You see the defects. You feel how bulky the packaging is. You learn what a day of 30 orders does to your back and your calendar. That's cheap education.
The trap is staying too long. Home storage has three ceilings and you'll hit one. Space runs out (a 5,000-unit reorder does not fit in a bedroom). Time runs out (past 20 to 30 orders a day you're a full-time packer, not a founder). And it wrecks your home and your family's patience. The move off home usually lands around 20 to 30 orders a day, sustained, not a one-off spike.
Stage 2: Rented self-warehouse (control, at a fixed cost)
The next rung is your own space. You lease a small godown, put up racking, and hire one or two people to receive stock and pack orders. In metro peripheries, grade B warehouse and godown space runs roughly ₹15 to ₹30 per square foot a month, with prime logistics parks costing more. So a modest 1,000 sq ft space is ₹15,000 to ₹30,000 a month in rent alone, before staff, electricity, racking and packaging.
Why choose this over a 3PL? Control. Your team, your packing standard, your custom unboxing, your stock counted your way. It suits brands with steady volume who care intensely about the box experience, or who have fragile or oddly shaped products a generic 3PL handles badly. The catch is fixed cost: rent and salaries hit whether you ship 500 orders that month or 5,000, where a 3PL flexes with your volume. So this rung makes sense when volume is high and steady enough to keep the space busy, but you'd rather own the operation than pay a partner's margin on top.
Stage 3: 3PL fulfilment centre (the default once you scale)
A 3PL (third-party logistics provider) stores your inventory in their warehouse and picks, packs and ships each order for you. You send stock once; they handle every parcel after. On warehousing, what you pay is storage plus per-order handling. Storage is billed one of two ways: per unit per month, or by the space you occupy (per cubic foot, or as an occupancy rate), which WareIQ explains here. Small, dense SKUs are cheaper on per-unit; bulky items are better on occupancy. Some partners soften the entry: Shiprocket Fulfillment charges no storage for the first 30 days a unit sits in its warehouse, per its pricing page, then storage applies.
Most brands land here past roughly 1,000 orders a month for one reason: the 3PL's cost flexes with your volume, you carry no lease, and you stop being a warehouse worker. You trade fixed cost and control for variable cost and freedom. The full when-to-switch logic and per-order maths sit in the 3PL guide; for the warehousing decision, the point is that someone else's roof holds your stock and you pay for space plus handling.
Stage 4: Marketplace fulfilment centres (FBA, Flipkart, Meesho)
The fourth place your stock can live is inside a marketplace. With Amazon FBA (Fulfilment by Amazon), you ship inventory into Amazon's warehouses; they store it, and when an Amazon order comes, they pick, pack and deliver it, and handle returns and customer service for that channel. Flipkart (F-Assured) and Meesho run their own versions. The pull is real: your listings get the Prime badge, faster delivery, and the buy-box edge that comes with platform fulfilment.
The catch is the cost structure, which founders underestimate most. FBA storage is cheap when demand is low and expensive exactly when you need it: off-season monthly storage is modest, but in the Oct-Dec festive peak, per-cubic-foot storage roughly triples as warehouse space gets scarce, per TheGSTCo's breakdown of Amazon India storage charges. On top sits the per-unit fulfilment fee, and every Amazon fee carries 18% GST. Treat this as a channel-storage decision, not your whole warehouse: use FBA for the Amazon slice (read how to sell on Amazon in India for the full picture), and never make it the only home for your stock, or your website orders get held hostage to peak-season rates.
The hidden costs that wreck the maths
The headline numbers are never the real numbers. Three hidden costs quietly turn a cheap-looking option expensive.
FBA aged-inventory penalties
Beyond the Oct-Dec spike above, the second FBA bite is aged inventory: stock that sits too long gets long-term storage surcharges on top of normal storage, and Amazon reviews aged inventory monthly rather than twice a year, per Amazon Seller Central India. Keep only fast movers in FBA, and pull or dispose of dead stock before the surcharge clocks pass the 180 and 365-day marks.
The 3PL SLA reality
A 3PL quote is a promise, not a fact. "Same-day dispatch" and "99% accuracy" look great in the deck, until the first festive rush when their warehouse is jammed with everyone's stock and your dispatch slips to day-2. You gave up direct control, so you now live on their SLA. Before you commit, send a small trial batch, audit picking accuracy and dispatch timing on your own orders, and confirm in writing how RTO returns are re-shelved, because a returned unit that isn't put back into sellable stock quickly is a phantom stockout.
The self-warehouse fixed-cost trap
Your own godown feels cheap per order when you're busy and ruinous when you're not. Rent, salaries and electricity don't care that this month was slow. A ₹25,000 rent plus two salaries is trivial per order on 3,000 orders, and brutal on 800 in a quiet month. Only take on a lease when your worst month still keeps that space reasonably busy.
Stuffing everything into FBA before a Diwali sale, then eating peak storage on what didn't sell. A founder ships 4,000 units into Amazon in early October to "be ready," chasing the Prime badge across the whole catalogue, slow SKUs included. Half sell. The other 2,000 units sit through Oct, Nov and Dec at roughly triple storage, and a chunk crosses the aged-inventory line into surcharge territory. Cheap platform storage in July became a five-figure bill across the peak quarter, plus removal fees to pull the dead stock. FBA is for fast movers, not your whole range. Send only what the data says will sell inside the peak.
Multi-location warehousing: the zonal shipping and RTO play
At real scale, a second decision appears: one location or several. Multi-location warehousing means splitting stock across two or more sites, say Delhi and Bangalore, so each order ships from the site closest to the customer. It's a genuine lever, but a scale move, not a starting point.
The upside is twofold. First, cheaper zonal shipping: a parcel from Bangalore to a Bangalore customer is a local-zone charge, far cheaper than shipping it down from Delhi on the national zone, so a big share of orders convert from expensive long-haul to cheap local delivery. Second, lower RTO: faster, more reliable delivery from a nearby warehouse means fewer failed handoffs and COD rejections. Since RTO hits margin directly, cutting it is real money; the mechanics live in reducing RTO on COD orders.
The cost is why you wait. Every location needs its own safety stock, so you freeze duplicate buffer inventory in each site, you need clean demand-by-region data to split correctly, and you run two pools to reconcile. Get it wrong and a bestseller stocks out in the south while it gathers dust in the north. The full go/no-go test is below.
Inventory Confidence Model™ (applied to multi-location): before you split stock across warehouses, answer four things. (1) Does my order data show a hard regional split, north vs south, not a vague hunch? (2) What extra safety stock must I hold in each location, and how much cash does that freeze? (3) Do the zonal shipping and RTO savings clearly exceed that frozen cash plus the added ops? (4) Can I reconcile two stock pools cleanly every week without a bestseller silently going out of stock in one? If any answer is shaky, stay single-location. A second warehouse you can't keep balanced is just dead stock in two cities instead of one.
The decision: match the warehouse to your volume, AOV and SKUs
Three numbers decide where your stock should live: orders a day, average order value (AOV), and SKU count. Volume tells you the rung, AOV tells you how much fulfilment cost you can absorb, and SKU count tells you how bulky your storage is.
| Your situation | Where stock should live | Why |
|---|---|---|
| Under ~20-30 orders/day, few SKUs | Home / garage | Free space, close to product, keeps cash |
| Steady 30+ orders/day, high AOV, custom unboxing matters | Rented self-warehouse | Control and packing quality justify the fixed cost |
| Past ~1,000 orders/month, want to stop packing | 3PL fulfilment centre | Variable cost, no lease, hands off your plate |
| Selling meaningful volume on Amazon | FBA for that channel + 3PL/self for website | Prime badge on Amazon, control your own D2C stock |
| High volume, clear regional demand split | Multi-location (add a 2nd site) | Cheaper zonal shipping, lower RTO |
If under ~20-30 orders/day → keep stock at home, invest nothing. If past that and overrun → choose a self-warehouse (you want control and custom packing) or a 3PL (you want variable cost and your evenings back); past ~1,000 orders/month the 3PL usually wins. If a real share of sales is on Amazon → put only fast movers in FBA and keep website stock in your 3PL or godown, never all-in on FBA. If your data shows a hard north-vs-south split and volume is high → add a second warehouse. If in doubt → stay one rung lower and cheaper; you can always move up when volume forces it.
Through Atomberg's hyper-growth phase I watched warehousing decisions get made emotionally, not on data. Founders sign a big godown lease because it feels like "a real company now," then pay rent on half-empty racks for six months. Or they dump everything into FBA for the Prime badge and get wrecked by peak storage on stock that didn't move. The discipline is boring: match the storage to your actual, sustained order volume, not the volume you hope for next quarter. Your stock's home is a cost centre, not a trophy. Revisit it every time your daily order count doubles.
- Write down today's real numbers: sustained orders/day, AOV, and SKU count.
- If you're under ~20-30 orders/day, stay home; don't sign anything yet.
- Before any lease, model your worst month, not your best, against the fixed rent and salaries.
- Get a 3PL quote on your exact order and SKU profile (storage plus per-order pick-pack), then compare that variable cost against a godown's fixed monthly nut.
- Put only proven fast movers into FBA; keep slow SKUs out to dodge aged-inventory surcharges.
- Diarise a stock review before Oct so you never carry dead FBA inventory through peak storage.
- Send a trial batch to any new 3PL, audit picking accuracy and dispatch timing, and confirm in writing how RTO returns are re-shelved.
- Revisit multi-location only when order data shows a hard regional split and volume justifies duplicate safety stock.
Next action: place your stock on the ladder today
Write three numbers: sustained orders per day, AOV, and SKU count. Find your row in the decision table above. That's where your stock should live right now, not where you hope it lives in a year. Under 20 orders a day, stay home and keep your cash. Past it and drowning, get one 3PL quote and one godown rent quote on your real numbers this week and compare the fixed nut against the variable cost. And never buy more stock than you can store cheaply and sell inside 90 days, because the most expensive warehouse in India is the one full of inventory that isn't moving. Once storage is sorted, how much to reorder and hold is the inventory management decision, and the full brand-build picture lives in how to start a D2C brand in India.
If you'd like the complete execution system, calculators, SOPs, templates and operating frameworks behind this process, continue inside D2C Acquisition.Lab.
