Inventory management for a small online store

For a small store, inventory management sounds like a warehouse problem. It is really a sales problem. Every time your store sells something you don't have, you have to message a buyer who is already expecting a parcel, apologise and cancel, and some of those buyers won't come back. The opposite mistake is quieter and just as costly: stock sitting on your shelf that your store shows as sold out, losing sales you never even see.

Fixing it doesn't need a warehouse system. It needs one count everyone trusts, a weekly habit of checking it, and a few decisions made in advance about what to do when stock runs low, runs out, or refuses to sell.

This guide is for a store run by one person or a small team, with stock on a shelf, in a spare room or in a godown. It covers tracking per variant, SKUs, counting and reconciling, reorder points, backorders and pre-orders, what to show shoppers, supplier gaps, dead stock, and bulk updates by spreadsheet.

Why stock accuracy is a sales problem

A wrong stock count turns into one of three problems, and each costs more than it looks:

• Overselling. A buyer orders the last Medium, which was actually sold yesterday on WhatsApp. You cancel, refund if they prepaid, and lose their trust along with the sale.
• Underselling. The store says sold out while two sit in a bag in the corner. Nobody complains, so you never find out what it cost.
• Sending the wrong thing. Mislabelled stock means a Large goes out instead of a Medium, and a sale becomes a return.

A lot of mismatches start the same way: one pile of stock, and several places selling it. Your store, your Instagram DMs, your WhatsApp, perhaps a marketplace or a stall. Every sale made outside the store is a sale the store doesn't know about.

The fix is to make your store's count the single source of truth and put every sale through it. When you take an order on WhatsApp or over the phone, record it as a manual order, so it comes off the same stock as an order placed on the store. If you also sell on a marketplace, set aside a fixed quantity for it and take that out of your store's count, rather than letting two systems sell the same units.

Track every variant, and give each one a SKU

"Ten kurtas in stock" is not a useful number when eight of them are Large. Buyers don't buy a product; they buy a size in a colour. So stock has to be counted at that level, every combination separately.

In Storecrafto, a product's sizes and colours become variants, and each variant carries its own stock count. The store knows it has three in Medium Black and none in Small White, and won't sell the one it doesn't have.

Then give every variant a SKU, a short code that identifies exactly one sellable item. It doesn't need to be clever. It needs to be readable and consistent, and a pattern of product, colour and size works well: KRT-BLU-M for a blue kurta in Medium. A few rules keep SKUs useful:

• One SKU per variant, never shared between two different items.
• Never reuse an old SKU for a new product, even after the old one is gone.
• Write the SKU on the shelf label, the bin or the bag, not only in the store.
• Keep your supplier's code next to your SKU, so a reorder asks for exactly the right item.

The physical side matters as much as the data. Keep each variant in one labelled place, and keep returns in a separate spot until they have been inspected. If you can't put your hand on a variant quickly, your count for it isn't reliable either.

Count and reconcile every week

A stock count is only useful if you compare it with what the store believes and fix the difference. Once a week, at a quiet time:

1. Count the physical stock of every variant, or one section a day so everything is covered weekly.
2. Compare each count with the stock figure in your store.
3. For every difference, find the reason before you change the number.
4. Correct the store's figure and write the reason in a simple log.

The reasons repeat: a DM sale nobody recorded, a return that went back on the shelf but not into the store, a damaged piece thrown away, a miscount when a delivery arrived. The log is where the pattern shows. If one product keeps drifting, something about how you handle that product is wrong, and counting alone won't fix it.

Do a full count before any big sale or launch, such as the weeks before Eid or before you run ads, because that is exactly when overselling does the most damage.

Returns deserve their own habit. When a refused or returned parcel comes back, inspect it before it goes near your shelf. If it is sellable, add it back to that variant's stock and check that the count really changed. If it isn't, write it off and log why.

Bulk updates with a spreadsheet

Typing corrections one variant at a time is fine for a handful of changes. After a full count or a big delivery from your supplier, a spreadsheet is faster. Storecrafto exports your products as a CSV file with one row per variant: update the stock column in Excel or Google Sheets and import it back. Rows that keep their product ID update that product instead of creating a new one.

A few habits make this safe. Try a small file first, spot-check a few variants in the store after importing, and keep the export from each count day, because it doubles as a record of what your stock was on that date.

Reorder points and low-stock alerts

A stock-out is usually not a surprise so much as a reorder placed too late. The cure is a reorder point for each product: the stock level at which you must order more so the new stock arrives before the old runs out.

The arithmetic is simple. Take how many you sell a day, multiply by how many days your supplier takes to deliver, and add a buffer for a busy week or a late delivery. An example with invented round numbers: if a variant sells about 2 a day and your supplier takes 10 days, you need to reorder when you are down to 20, or around 25 with a buffer. Reorder at 10 instead, and you are sold out for half the wait.

Lead times move with the season and with your supplier's own stock, so revisit your reorder points whenever a delivery arrives late.

Storecrafto's low-stock alerts flag items that are running low, and your dashboard lists them in one place. Use that list as the trigger to check each item against its reorder point, and reorder anything at or below it the same day. An alert you read and then park is just a notification.

For clothing, the reorder is also a size decision. Your sales by variant show your real size curve, meaning which sizes actually sell, and the next order should follow it rather than buying the same number of every size.

Backorders and pre-orders: be exact about the promise

Selling past zero can be a sound decision. If you make to order, or your supplier restocks reliably within a known time, turning buyers away because the shelf is empty today loses sales you could have filled.

In Storecrafto it is a setting on the product that lets it keep selling at zero, explained in managing inventory and backorders. Make it a deliberate choice for each product, not a habit. It is a promise to deliver something you don't have yet, so treat it like one:

• Only turn it on for products you can genuinely restock, with a lead time you know.
• Say it on the product page: when it will ship, in plain words, not small print.
• Repeat the date when you confirm the order, so nobody is surprised.
• If the date slips, tell the buyer before they ask, and offer a refund if they would rather not wait.
• Leave it off for one-off, limited or seasonal items you can't get again.

Pre-orders are the same promise with a longer wait, and on cash on delivery the wait is the risk: the longer a buyer waits, the more time they have to change their mind before the rider arrives. For a long wait, it is reasonable to ask for payment upfront, for example by bank transfer, and keep COD for stock you can ship now.

What to show shoppers

Stock levels on a product page are a real lever, and they are only honest if the count behind them is right.

Storecrafto offers three ways to show stock on your store: the exact number in stock, a number that appears only when stock is low, or no number at all, just in stock or out of stock. The options are explained in how stock appears to shoppers. Each has its reason:

• Exact counts are transparent, but they also tell competitors how much you hold, and make slow sellers look slow.
• Low stock only shows a number when it matters. "Only 3 left" is real urgency when it is true.
• Hidden keeps your numbers private, at the cost of that urgency.

The line not to cross is invented scarcity. "Only 2 left" on something you have two hundred of, or a countdown that restarts every time the page loads, works once per buyer. After that, nothing on your store is believed. Real low stock, shown honestly, is persuasive enough.

When your supplier runs out

Sooner or later a supplier tells you that the thing you have been selling isn't coming. What matters then is how quickly you act:

1. Set the affected variants to their true stock straight away, and switch off selling past zero.
2. List every open order waiting on that item.
3. Message each buyer before they message you, with the honest position and a choice: wait, swap or cancel.
4. For anyone who cancels, cancel the order and return any prepaid money through your gateway or in cash.
5. Update the product page. A clear back-in-stock date, or a plain "no longer available", beats a silent sold-out badge.

If buyers mostly reach you on WhatsApp, that is where these messages belong, and WhatsApp for online sellers covers how to write them.

For your best-sellers, don't depend on a single supplier. A second source you have already tested, even at a slightly worse price, is insurance for exactly this week. If your supplier ships each order for you, the same thinking applies, and dropshipping in Pakistan covers the rest.

Dead stock and clearance

Dead stock is stock that has stopped selling: cash you have already spent, sitting on a shelf and taking up space you could use for something that moves. Choose a window that fits what you sell, a season for fashion and longer for everyday products, and treat anything that hasn't sold within it as a problem to solve rather than one to wait out.

Your analytics show what is selling and what is stuck. From there, roughly from gentlest to most drastic:

• Fix the listing: better photos, real measurements, an answer to the question buyers keep asking.
• Put it in a bundle with something that sells, so it moves at a price that still makes sense.
• Give it a clearance category and a discount code for your WhatsApp or email list.
• Price it to clear. The money you paid is gone either way, and cash in hand beats stock on a shelf.

If you use bundles, remember that in Storecrafto a bundle carries its own stock count, and selling a bundle doesn't reduce the stock of the products inside it. Count both, and adjust both.

Then learn from it. Dead stock often traces back to one decision: a size bought too deep, a colour nobody wanted, a trend that passed. The cheapest cure is to buy shallow and reorder fast, which is exactly what reorder points make possible.

The short version

• Every stock mismatch becomes a cancellation, a lost sale or a wrong parcel. Accuracy is a sales problem.
• Make your store's count the one source of truth, and record WhatsApp and phone sales in it.
• Track stock per variant, with a readable SKU for each.
• Count weekly, find the reason for every difference, and log it.
• Set a reorder point per product from daily sales, supplier lead time and a buffer.
• Sell past zero only with a restock date you trust, stated plainly to the buyer.
• Show real scarcity or none. Invented scarcity costs you trust.
• Clear dead stock deliberately, then buy shallower next time.