Multi Location Inventory Management: Beyond the Warehouse Grid
Multi location inventory management guides assume every location is a warehouse or store. The hard version is vans, hospital shelves, client sites and forward stocking locations. What breaks there, and what a system needs per location type.
When a location is a van, not a warehouse
Search for multi location inventory management advice and you get the same picture everywhere: a neat grid of warehouses and stores, each with staff, shelving, Wi-Fi and a receiving dock. Real distributed operations look nothing like that. The locations that matter are engineer vans, hospital shelves, consignment points at customer sites and forward stocking locations in industrial parks. No staff. No fixed infrastructure. Often no signal. The standard playbook breaks there, and so does most software.
Ventory is a field inventory platform built for exactly these locations: the vans, client sites and forward stocking locations that sit between an ERP and the people doing the work. This article covers what breaks at those locations, and what a system needs per location type.
Why multi location inventory management fails in the field
The stakes are not small. IHL Group's September 2025 study puts the global cost of inventory distortion, out-of-stocks plus overstocks, at 1.73 trillion dollars a year, 6.5% of retail sales, despite 172 billion dollars spent on improvements in a single year. Money keeps going into systems, and the distortion persists. One reason: the fixes concentrate on the locations that are easy to instrument, the warehouses and stores, while the hardest stock sits elsewhere.

In field operations the pattern is sharper. A part that shows "in stock" in the ERP is actually in a van two regions away. A technician arrives without the part, the fix fails, and a second visit is booked. Consigned stock at a customer site quietly expires because nobody counts it. Ghost stock accumulates until the annual count writes it off. Ventory exists because ERPs record what shipped to these locations, not what is really there; that gap is what field inventory management closes.
Root causes: what actually breaks outside the warehouse
- Counts stop happening. Cycle counting assumes staff at the location. Nobody flies an auditor to 300 vans, so accuracy decays until the year-end shock.
- Transfers go unrecorded. An engineer grabs a part from a colleague's van. The job gets done, the system never hears about it, and two van records are now wrong.
- Replenishment runs on guesswork. Without live consumption data per location, planners push stock on gut feel. The result is both overstock and stockouts at once.
- Ownership blurs. Consigned stock, customer-owned stock and own stock share shelves. When ownership is unclear, so is the P&L.
- Connectivity is assumed. Hospital basements, substations and moving vehicles drop signal. Any workflow that needs a live connection silently stops being used.
What a system needs, per location type
Treating every location the same is the core mistake. Each location type fails differently and needs different controls. Before any software decision, two basics apply everywhere. First, every location needs a unique identity in the system, down to the individual van or shelf, with a naming convention that survives staff turnover. Second, every item movement needs a scan, because typed entries decay into guesses within weeks. With those in place, the per-type requirements look like this.
Engineer vans and trunk stock
The location moves and the operator is alone. The system needs scan-based issue and return in seconds, per-van min-max levels, and full function while offline, syncing when signal returns. Counts must be possible in the five minutes before a shift, not on a depot recall. More on this in our guide to trunk stock management.
Customer sites and consignment points
Here the problem is ownership and proof. The system must track consigned stock by owner, capture proof of use at the moment of consumption, and manage expiry dates so product rotates before it becomes a write-off.
Forward stocking locations
Unstaffed by design, shared by many. The system needs identity on every transaction (who took what, when), alerting when levels cross thresholds, and rule-based replenishment from the central warehouse so the location refills itself.
Hospital shelves and points of use
Regulated stock with lot and serial numbers, handled by clinical staff who will not learn logistics software. Scanning must capture lot, expiry and UDI data in one gesture, and instant cycle counting must fit into minutes between cases.
Across all four types, one requirement repeats: two-way sync with the ERP. SAP, Oracle, Dynamics, Sage or NetSuite stays the system of record; the field layer feeds it the truth from each location. Transfers deserve special attention. In the field, the fastest source of a part is often another location: the nearest van, the next site over. A good system makes a location-to-location transfer a two-scan action, so the shortcut that engineers will take anyway becomes a recorded movement instead of a silent discrepancy.
Proof: 450+ field locations, one live view
A global medtech manufacturer runs Ventory across 450+ field locations: hospital consignment points, loaner kits and regional stocking locations. Field teams scan on their phones, counts happen where the stock sits, and every movement syncs to the ERP. The company achieved a +70% reduction in stockouts and +35% excess stock reduction across the network.
A national ambulance service applies the same model to stations and vehicles and holds 99.76% rolling stock accuracy, measured continuously rather than at an annual count. Neither organisation added warehouse infrastructure. They gave the people already at each location a tool that takes seconds to use.
The common thread is where accuracy comes from. It is not a central team checking harder. It is hundreds of small, frequent transactions captured at the point of use, each one too easy to skip. Multiply that across every van and shelf, and the network stays accurate by default.
Getting started: five steps
- List every real location. Vans, site rooms, lockers, shelves, kits. If stock sits there overnight, it is a location.
- Classify by type, not size. Staffed or unstaffed, owned or consigned, fixed or mobile, connected or offline. The controls follow the type.
- Set stocking rules per location. Min-max per SKU per location, based on actual consumption where known and best estimate where not.
- Put scanning in the hands already there. Engineers, nurses, drivers. If a transaction takes more than a few seconds, it will be skipped.
- Connect the ERP and measure. Track stock accuracy, stockouts and excess per location type. Fix the worst type first, then expand.
Start small on purpose. One region, one location type, a 90-day baseline. The accuracy delta from that pilot is the business case for the rest of the network, and it usually writes itself.
Field notes
- Inventory distortion costs 1.73 trillion dollars a year globally, 6.5% of retail sales, even after 172 billion dollars of fixes (IHL Group, 2025).
- A global medtech manufacturer runs 450+ field locations on one platform and cut stockouts +70%.
- A national ambulance service holds 99.76% rolling stock accuracy across stations and vehicles.
Managing stock across vans, client sites and stocking locations? Get in touch with the Ventory team.
Frequently asked questions
What is multi location inventory management?
It is the practice of tracking, counting and replenishing stock across more than one physical site from a single system. The hard version involves locations without staff or infrastructure: vans, customer sites, forward stocking locations and hospital shelves, where standard warehouse tooling does not apply.
Why is my ERP not enough for multiple field locations?
An ERP records planned movements: what was shipped, issued or invoiced. It does not see unrecorded transfers, field consumption or decay at unstaffed locations. A field inventory layer captures those events at the point of use and syncs them back, so the ERP reflects reality.
How do you count inventory at locations with no staff?
Use the people who visit anyway. Engineers, drivers and reps run scan-based counts in minutes on their phones, guided by the app. Frequent small counts at the point of use replace the annual audit, which is how a national ambulance service maintains 99.76% rolling stock accuracy.
What if locations have no internet connection?
The system must work fully offline: scanning, counting, issuing and returning without signal, then syncing automatically when the device reconnects. If offline mode is missing, transactions in basements, rural sites and vehicles simply never get recorded.
How do you handle consigned stock across many customer sites?
Track owner, lot and expiry per item, and capture proof of use at the moment of consumption. That turns consignment from an untracked liability into auditable stock. It also keeps invoicing clean, because usage is recorded when it happens rather than reconstructed at the quarterly count.
How long does it take to roll out across hundreds of locations?
With a field inventory layer, typically live in 2-3 months, because there is no hardware to install per location and training takes minutes per user. Rollouts scale by loading locations and users, not by site projects.
About Ventory
Ventory is the field inventory layer for regulated, high-stakes industries. We give MedTech, 3PL, Aerospace, Energy and FMCG leaders real-time visibility and control over inventory outside the four walls, in hospitals, ambulances, trunk stock, consignment locations, and field service vans. Ventory is ERP-agnostic (SAP, Oracle, Dynamics, Sage, NetSuite) and trusted by a global medtech manufacturer, a national ambulance service, global logistics and consumer-goods operators. See how it works →
