Solutions · Inventory & Order Management
Stock counts your team trusts, orders that route themselves, and receiving, picking and returns that follow one set of rules. We build inventory and order systems around how your warehouse, storefronts and sales channels actually work.
You own the code and IP Weekly demos Reply within one business day
Basket · 3 items
Optimised checkoutMerino crew
Canvas holdall
Ceramic carafe
Merino crew · medium
Canvas holdall
Ceramic carafe · pair
Subtotal · delivery included
2.9%
Conversion
41s
To checkout
−22%
Cart abandonment
OlDevs is a full-stack technology studio in Vancouver, British Columbia that builds custom inventory and order management systems.These systems hold one accurate stock record across locations and sales channels, then drive purchasing, receiving, barcode picking, order routing, fulfilment and returns from it. They connect to e-commerce platforms, accounting ledgers and shipping carriers so operations and finance read the same numbers. We build them when an off the shelf product cannot match the way a business really moves goods.
Key facts
Stock you can count on
Stock lives in one ledger covering warehouses, storefronts, bins, consignment and goods in transit. Availability is calculated from on hand, reserved and inbound quantities, with per channel buffers so a single unit is never sold twice.
Purchase orders carry supplier lead times, minimum order quantities and expected costs. Receiving handles partial deliveries, over shipments, damaged goods and backorders, and matches packing slips to orders and invoices before anything hits the ledger.
Handheld scanners, phones and label printers replace typing. Staff scan to receive, pick, pack, transfer and count, with lot, serial and expiry capture where you need it, plus cycle counting that corrects drift without shutting the floor down.
Allocation rules decide which location fills each order, when to split a shipment and when to hold. Pick lists, packing checks, carrier rates and labels follow the same rules, so promises made at checkout match what leaves the dock.
Returns arrive with a reference, get inspected, then restock, repair or scrap on a recorded decision. Refunds, exchanges and credit notes post back to the order and the accounting ledger, so nothing is written off in a spreadsheet nobody reviews.
Connections to e-commerce platforms, accounting software, point of sale, carriers and EDI partners run as queued, retryable jobs. Every sync leaves a record, and reconciliation reports show exactly where two systems disagree and why.
Process
Walk the floor
We watch how goods actually arrive, move and ship, including the workarounds. Exceptions, not the happy path, decide whether a custom build is worth it.
Fix the data first
Item master, units of measure, locations, suppliers and channel mappings get cleaned and agreed before code. Bad master data ruins good software.
Build in weekly demos
One accountable team ships a working demo every week, starting with receiving and counting so staff can use it early and tell us what is wrong.
Pilot in one location
The new system runs alongside the old one at a single site. We reconcile counts, train staff and settle the rules before anyone else is affected.
Cut over and extend
A short freeze, a full count, then go live with monitoring in place. Other locations, channels and integrations follow once the first site is calm.
How we work
Quantities are never typed over. Receipts, picks, transfers, adjustments and returns are recorded as dated, attributed movements, and the count you see is the sum of them. That is what makes a real audit trail possible.
Sellable stock comes from on hand minus reserved plus confirmed inbound, adjusted by channel buffers. Marketplaces and your storefront get the number they should see, not a raw shelf count.
Barcodes on items, bins and documents mean fewer keystrokes and fewer wrong SKUs. Where barcodes are missing, the system prints them, so the physical world and the database use the same identifiers.
Allocation, substitution, backorder and hold logic sit in a rules layer your managers can read and change, rather than being buried across integrations where nobody can tell why an order went to the wrong site.
Every sync is idempotent, queued and retried, with a dead letter queue a human can inspect. A carrier or platform outage delays a message; it does not silently lose an order or double a shipment.
Stock valuation, shrinkage, ageing, supplier performance and fill rate all read the same movement records the floor creates. Finance and operations then argue about decisions, not about whose export is right.
Who it's for
Multi site operations that need one stock picture across warehouses, plants and retail, with permissions, approval limits and an audit trail internal auditors and external accountants can follow.
Program inventory, asset pools, publications and equipment loans, tracked with accountable custody, bilingual EN and FR interfaces where required, and WCAG 2.2 AA accessibility built in.
Head office visibility over stock and orders at every location, with ordering rules, transfers between sites and reporting that respects what each franchisee is allowed to see and change.
Growing brands that have outgrown spreadsheets and channel apps but are not ready for a heavy platform. We start with the parts that hurt and add the rest as volume justifies it.
Add figure
Orders routed per day
Add figure
Stock accuracy after pilot
Add figure
Locations live on one system
FAQ
Buy, if your operation looks like everyone else's: standard SKUs, one or two locations, simple picking and a common sales channel. Products in that space are mature and far less work to adopt than to rebuild. Build when your routing, kitting, lot rules, contract terms or approvals are the actual business and no product handles them without painful workarounds. We will tell you honestly which side you are on.
Usually yes. Platforms such as Shopify, WooCommerce and common accounting packages publish documented APIs, and shipping carriers publish rating and label services. We confirm what each API supports at the time of the build, since capabilities and limits change, and we design the sync so a platform outage delays messages rather than corrupting your stock records.
Often not at first. Modern phones and tablets scan well enough for receiving and counting, and a basic label printer covers most needs. Dedicated scanners earn their place in high volume picking, cold storage or environments where gloves and drops are normal. We test with the hardware you already own before recommending a purchase.
They are captured at receiving and carried through every movement, so you can trace a unit from supplier to customer and back on a return. Picking can enforce first expiry first out, and recalls become a query rather than an archaeology project. Where your sector has specific record keeping rules, we build to what your compliance team confirms. This page is not legal advice.
That is the point of the movement ledger. Every change has a date, a person, a reason and a document behind it, and valuation reports are derived from those records rather than a separate spreadsheet. We involve finance early so the chart of accounts mapping, costing method and cut off rules are agreed before go live, not argued about after.
We import the item master and open orders first, run the new system in parallel at one location, and reconcile daily until the counts match. Cutover is then a short freeze, a full physical count and a switch, usually over a quiet weekend. The old system stays readable afterwards so history is never lost.
You do. Clients own all code, designs, accounts and IP, and we hand over repositories, documentation and infrastructure access as part of delivery. You can keep us on for support and further work, move it to your own developers, or do both. Nothing in the build is designed to make leaving difficult.
Let’s connect
Tell us what you’re building. We’ll reply within one business day with next steps and a tailored quote — no obligation.