Why Does Purchasing Keep Over-Ordering What You Already Have Somewhere Else?

Every multi-location business has a version of this story. A site runs low on a part, someone raises a purchase order, and the supplier ships it. At another site, the same item has been sitting on a shelf for months, untouched and fully paid for. Nobody did anything wrong. The system each person could see told them to buy.

That is the over-ordering inventory problem in its most common form. It is rarely a discipline or forecasting failure. It is a visibility failure: purchasing decisions are made site by site, against local numbers, while surplus elsewhere in the network stays invisible. Adding locations makes it worse, because every new site adds another buffer and another blind spot.

This article covers the causes, how to diagnose the problem in your own network, what it really costs, and the steps that reduce it. Then it explains how The Ninjaz Group approaches it, and where that approach has limits.

What Actually Causes Excess Inventory Across a Network

Most excess inventory causes trace back to a handful of patterns. Almost all of them intensify as you add sites:

• Local reorder points. Each site sets its own minimums to avoid stockouts. The network ends up carrying the sum of every site's safety buffer instead of one shared buffer.
• No view of surplus elsewhere. The buyer sees on-hand stock for their own location only, or sees a figure from another system that is out of date.
• Lead-time padding. Buyers who have been burned by late deliveries order early and in larger quantities, and nobody ever reverses the habit.
• Supplier minimums and price breaks. Order quantities follow pack sizes or discount tiers rather than network demand.
• Stock that exists but isn't visible. Returns, project leftovers, and transfers in transit sit in a status that buyers' reports ignore.
• Duplicate item records. The same part lives under different codes at different sites, so even a diligent search for it comes up empty.

Notice that none of these require a careless buyer. Each is a reasonable local response to incomplete information. That is why asking purchasing to "be more careful" doesn't work. The information they need is not in front of them at the moment they commit money.

How Fragmented Systems Turn a Reasonable Order Into a Duplicate One

Consider a group with twelve locations. Finance and procurement run on one ERP, but inventory is tracked differently across the group. Some sites use a warehouse management system, a few inherited from acquisitions run on spreadsheets, and the rest rely on whatever the ERP's inventory module shows. Each system is "correct" in its own world.

A project at Site 4 is cancelled, leaving forty units of a common component on a shelf. The units are counted, but they are tagged to a closed project, so they don't appear in available stock. Three weeks later, Site 7 runs low on the same component. The buyer checks local on-hand, finds almost nothing, and raises an order for thirty units because that is where the supplier's price break starts.

The group now holds seventy units of something it needed thirty of, and Site 7's order is already on a truck.

Transferring from Site 4 was technically possible. But it would have meant knowing the stock existed, finding someone to ask, agreeing on who pays for freight, and getting the units moved. Ordering from a supplier takes a few clicks. When buying is easier than finding, buying wins every time, and across twelve sites and thousands of items it happens constantly.

How to Diagnose It in Your Own Network

You don't need a transformation program to find out how bad it is. You need a few weeks and some honest data pulls. Work through these steps:

1. Pull on-hand by site for your top fifty items by spend. Put it in one table. If assembling that table takes days, you have already learned something.
2. Compare days of supply to consumption at each site. Flag sites holding far more than they use and sites running thin. In most networks you will find both, for the same items, at the same time.
3. Match open purchase orders against surplus elsewhere. Count how many open POs are for items another site holds above its own need.
4. Audit the item master for duplicates. Search for the same description under different codes, units of measure, or suppliers.
5. Trace a sample of recent POs. For each, ask what the buyer could actually see when they ordered. Was network stock visible, or only local stock?
6. Compare transfers to purchases. If sites almost never transfer stock to each other, the process is probably too hard to use, not unnecessary.

The most telling result is usually step three. A list of live orders that duplicate stock you already own is hard for any executive to ignore.

The Real Cost of Over-Ordering

The obvious cost is cash tied up in stock that isn't moving. But the carrying cost is only the start.

Inventory and cash. Surplus occupies shelf space, insurance, and handling time, and it risks obsolescence or damage while it waits. Money spent on duplicates is not available for stock you do need.

Time. Staff spend hours reconciling counts, chasing "where is it?" questions, and processing returns or write-offs that better purchasing would have avoided.

Lost deals. Overstock and stockouts coexist. While one site is buried in a part, another turns a customer away or delays a job because it ran out. The sale is lost even though the network owns the item.

Mistakes. When records disagree, people work from workarounds: a private spreadsheet, a message to a colleague, a guess. Wrong parts get ordered, wrong quantities get received, and corrections eat more time.

Burnout. Site managers and buyers are blamed for outcomes the system caused. Constant firefighting over stock they can't see is tiring and corrosive.

Customer frustration. Delayed repairs, backorders, and unreliable promises damage trust, especially when customers can tell the business is disorganised behind the scenes.

Steps That Reduce Overstock

If you want to know how to reduce overstock without replacing every system, work on the decision point first: the moment a buyer commits to an order. Here is where each part of your stack contributes.

ERP. Establish a single item master with one code per part and consistent units. Make sure on-hand quantities can be queried across all sites, with clear rules about which statuses count as available. Stock tagged to closed projects, returns, or quarantine should be reviewable, not buried.

Procurement. Add a check before a purchase order is released: does any other site hold this item above its own need? For orders over a set value, require the buyer to confirm that network stock was reviewed. Consolidate recurring purchases across sites so you reach supplier price breaks once for the group, not once per location.

Inventory management. Move from independent site minimums toward network-level targets, with allocation to sites. Classify slow-moving and surplus items explicitly so they stop looking like healthy stock.

Warehouse management. Accurate receiving, put-away, and bin-level records are what make any cross-site view trustworthy. Cycle counts and prompt recording of returns and transfers keep the numbers honest.

Supplier portals. Share consolidated forecasts and confirm real lead times, so buyers have less reason to pad orders. Use portal data on delivery performance to set buffers deliberately. Ask suppliers about return or buy-back terms for surplus.

BI and reporting. Build a weekly view of open POs against surplus elsewhere, days of supply by site, and transfers versus purchases. Reporting that is reviewed on a schedule changes behaviour. Reporting that nobody opens does not.

What Network-Wide Visibility Changes About Purchasing

The biggest shift is in the question the buyer asks. Instead of "do we have this here?", it becomes "do we have this anywhere, and what is the cheapest way to get it where it's needed?"

That reframing opens up options that local-only buying never offers. A requirement can be met by transferring surplus, redistributing stock before it ages, pooling several sites' needs into one order, or buying new only when none of those make sense. In the earlier scenario, Site 7's buyer would see Site 4's forty units, with cost, location, and status, and could request a transfer instead of placing an order.

Visibility also changes who is accountable. When stock is seen as a network asset, purchasing can own network-level outcomes such as turns, surplus value, and duplicate orders avoided, rather than being measured only on whether each site stayed in stock.

There are honest limits. Visibility doesn't remove freight costs, transfer pricing between legal entities, or the judgment about whether moving a low-value item is worth it. It also depends on records being accurate. Showing a buyer wrong numbers across the whole network is worse than showing them wrong numbers at one site. Cross-site visibility works only when counts, statuses, and item records are maintained by someone who owns them.

How The Ninjaz Group Approaches It

The Ninjaz Group is a managed-services partner for multi-location enterprises, and this problem sits squarely in its territory: the stock data exists, but it lives in separate systems that don't talk to each other.

An engagement starts with a Blueprint. We map how the operation works, including where stock records live, what buyers look at when they order, and where work is done twice, data goes stale, or money leaks. You keep the map whether or not you continue with us.

Next comes Build & Link. There is no rip-and-replace: the ERP, WMS, and inventory tools you've already paid for stay in place. Through our Managed Link layer, we build, own, and maintain the integrations between them. The Ninjaz Operating Platform is the hub that makes this work, with connectors to the ERP, inventory and WMS, finance, and other systems you already run, plus a unified data layer that holds one record per asset and order.

For this problem, the relevant pieces are:

• Inventory and asset views across sites, so a buyer or planner sees stock beyond their own location.
• A Command Center with live KPIs and alerts across locations, such as surplus value by site or open orders for items held elsewhere.
• Workflow automations that can route a purchase request for review, or prompt a transfer, when surplus exists elsewhere.
• Reports, roles, and an audit trail, so you can see who ordered what, based on what information.

After go-live, Managed Business Operations covers the day-to-day running, so the integrations and data stay healthy rather than drifting.

To be plain about the limits: the platform does not replace your ERP, and it can't fix poor counting discipline, unfavourable supplier terms, or unclear transfer rules. It gives your people one reliable view and a workflow around it. Your policies still decide what they do with it.

Start With the Orders You Shouldn't Have Placed

The quickest way to see whether this is costing you is to run the diagnostic above and list the open orders that duplicate stock elsewhere in the network. If the list is short, your processes are working. If it isn't, you have found the clearest case for connecting your systems.

If you'd like help mapping where your stock data lives and what it would take to connect it, book a consultation. If you'd rather see how cross-site inventory views and the Command Center look in practice, explore the platform demo.