Solution guide

Inventory and order systems for businesses that have outgrown spreadsheets.

A stock sheet works until two people sell the last item at the same time. As products, locations and orders multiply, the cost of not knowing what you have — overselling, over-ordering, dead stock — starts to exceed the cost of a system that does know.

Request a project quote Estimate the size of the project

The core of the system

Products and variants — items, sizes, colours, units of measure, bundles.
Stock by location — warehouse, shop, consignment, in transit.
Purchasing — purchase orders to suppliers, expected arrival, goods received.
Sales orders — from whatever channel they arrive: shop, website, phone or chat.
Movements and adjustments — every change to stock recorded with who, when and why.
Alerts and reports — what is low, what is slow, what is on its way.

The rule that holds it together: stock changes only through recorded movements. If anyone can overwrite a number, the numbers stop being trusted.

Be careful not to build an ERP by accident

Inventory projects grow in one direction: towards everything. Accounting, payroll, manufacturing and HR are different problems with mature products of their own. A good scope draws a firm boundary:

Keep accounting in your accounting software and connect to it.
Start with one flow — for example receiving stock and selling it — rather than every process at once.
Decide what the system is the single source of truth for, and what it merely reads.

For simpler needs, an existing inventory product may be enough. We would tell you if it is.

Hong Kong trading realities

Several currencies. Buying in RMB or USD and selling in HKD is routine. Decide how exchange rates are applied and when costs are fixed.
Orders arriving by chat and phone. Many orders still begin as a message. Make entering one quick, and keep the message and the order linked.
Bilingual product names. Products often need English and Chinese names, and staff search in both.
Several locations and cross-border movement. Shops, godowns and mainland partners each need to be a clear place for stock, with transfers recorded between them.
Mobile stock-taking. Phones can scan barcodes or QR codes, which makes counts faster and less error-prone.

Settle these before the first import

What is one "product"? Sort out variants and bundles first; it shapes everything.
How is the starting stock established? A physical count before launch avoids inheriting spreadsheet errors.
Who may adjust stock, and what must they record?
How do orders from each channel enter the system?
What flows to accounting, and when?

Effort grows with locations, product complexity, channels, integrations and reporting. The cost estimator is a good first check.

Common mistakes

Importing messy product data as it is, then wondering why reports are wrong.
Skipping a physical count at the start.
Allowing free-form stock edits "just this once".
Building features for processes nobody has defined.
Not training the people who handle stock every day.

An illustration: a shop and a godown

Imagine a trading business with a small shop and a godown. Stock is counted on a shared sheet, orders arrive by WhatsApp and phone, and nobody is sure which of the two places holds the last few units of a popular item. (An illustrative scenario, not a client.)

A focused first release would cover receiving stock and selling it at the two locations, one barcode per variant, recorded transfers between shop and godown, and a low-stock list. Accounting would stay exactly where it is, with invoices passed across. Everything else — returns, bundles, forecasting — would wait.

How to tell whether it worked

Choose a few plain measures before launch and look at them a month later:

How far physical counts differ from the system's numbers.
How long it takes someone to find out whether an item is available.
How many orders were delayed or cancelled because the stock turned out not to exist.
How much time staff spend reconciling figures instead of serving customers.

The aim is not a perfect number on day one. It is a number that becomes more trustworthy every week because every change is recorded.

Frequently asked questions

Do we need custom software, or would an existing inventory product do?

Often an existing product is enough. Custom software makes sense when your process is unusual, when several channels or systems must be tied together or when you pay for much more product than you use. We will give you an honest view.

Can it connect to our accounting software?

Usually, yes. We would agree what is shared — for example invoices and purchase costs — in which direction, and how mismatches are handled.

Can staff scan barcodes with their phones?

Yes, a phone camera can read barcodes or QR codes. Whether that is enough or you need dedicated scanners depends on volume and environment.

Can it handle several shops and a warehouse?

Yes. Each location is a place that holds stock, with transfers recorded between them. The more locations, the more important clear rules become.

How do we move our current stock sheet across?

We would clean and map the data, import it in stages and recommend a physical count to confirm the starting figures, then run alongside the old sheet for a short period before switching.

Not sure what you actually have in stock?

Tell us how stock and orders move through your business today. We will help you scope the smallest system that fixes the biggest problem.

Request a project quote