Vendor managed inventory turns replenishment on its head. Rather than you watching stock and raising orders, the supplier watches your stock and keeps it between agreed levels. Done well it cuts stockouts, removes a pile of low-value admin and gives the supplier a far clearer view of real demand. Done badly it creates dependency and arguments over stock nobody wanted. This guide covers how VMI works, how it differs from consignment, what belongs in the agreement, the risks, the KPIs and how to set one up.
Key takeaways
- In VMI the supplier monitors your stock and replenishes to agreed minimum and maximum levels.
- VMI is about who replenishes; consignment is about who owns the stock. Decide each separately.
- The arrangement lives or dies on data quality, not on goodwill.
- Start with one supplier and one high volume, steady demand category, then extend.
What is vendor managed inventory?
Vendor managed inventory is a replenishment model in which the supplier, not the buyer, decides when and how much to ship. You give the supplier visibility of your stock position and your consumption, you agree the levels that stock should stay between, and the supplier takes on the job of keeping it there. Purchase orders do not disappear, but they change character: they become a confirmation of what the supplier has already worked out, or a blanket agreement that individual deliveries are drawn against.
The logic is straightforward. The supplier knows their own lead times, production schedule and capacity far better than you do, and once they can also see your usage, they are the best placed party to plan replenishment. In a traditional model that knowledge is split across two organisations that talk only when an order is raised, which is exactly when it is too late to plan. VMI closes that gap by moving the decision to where the information is richest.
It is not a new idea. Retailers and their large consumer goods suppliers have run VMI for decades, and it has since spread into manufacturing, healthcare and facilities management. What has changed is the cost of entry: sharing accurate stock data once required a serious integration project, and now a reasonable procurement platform can export the same figures on a schedule.
How VMI works in practice
The mechanics are simpler than the concept suggests. Both parties agree, item by item, a minimum level below which stock must not fall and a maximum level above which it must not rise. The minimum protects against stockouts and reflects lead time plus a safety buffer; the maximum protects your shelf space, cash and tolerance for obsolescence. Between those two lines, the supplier has freedom to decide when to ship and in what quantity, which is precisely what lets them batch deliveries efficiently.
A typical cycle runs like this. On an agreed rhythm, daily or weekly, you send the supplier your current on-hand quantities and recent consumption. The supplier compares that to the agreed levels, factors in what is already in transit and their own lead times, and creates a replenishment proposal. Depending on how much autonomy you have granted, that proposal either ships directly or waits for a quick approval on your side. Goods arrive, you receive them against the agreement, and the cycle repeats.
Two practical details make or break the routine. The first is exceptions: a promotion, a new production line, a shutdown or a lost customer all break the pattern, so the agreement needs a named person on each side to flag them in advance. The second is receipting. VMI reduces ordering admin but does not remove goods receipting, and if deliveries are booked in casually the stock figures you send back degrade within weeks.
VMI does not mean nobody is watching. The single most common failure is treating the handover as the end of the work. Someone on the buying side still needs to review levels, challenge replenishment decisions and check that the data being shared is true. VMI moves the effort; it does not delete it.
VMI versus consignment stock
These two terms are used interchangeably far too often, and the confusion causes real disputes. They answer different questions. VMI answers "who decides when to replenish". Consignment answers "who owns the stock while it sits in my building". A supplier can manage your replenishment while you buy every delivery on arrival, and a supplier can own stock on your shelf that you order yourself in the traditional way.
| Model | Who raises replenishment | Who owns stock on site | Typical use |
|---|---|---|---|
| Traditional purchasing | Buyer | Buyer, from delivery | Most categories, especially variable or one-off demand |
| VMI | Supplier | Buyer, from delivery | Steady, high volume items with a trusted supplier |
| Consignment | Buyer | Supplier, until consumption | High value or slow-moving spares the buyer will not pre-fund |
| VMI with consignment | Supplier | Supplier, until consumption | Mature partnerships with strong data sharing |
Combining the two is the most attractive arrangement for the buyer and the most demanding for the supplier, because the supplier funds the stock and carries the planning risk. Expect to pay for that somewhere, in unit price or in a minimum volume commitment. It is a reasonable trade, but it should be a conscious one.
The data the supplier needs, and how it is shared
VMI is a data arrangement wearing a logistics costume. The supplier cannot plan what they cannot see, and the quality of what you share sets the ceiling on how well the model performs. Four things are essential: current on-hand quantity by item and location, consumption over a recent and consistent period, open orders and goods already in transit, and forward-looking changes you know about but the numbers do not yet show.
How that data travels matters less than whether it is right. Large partnerships use EDI or API integration; plenty of successful arrangements run on a scheduled file export or a shared portal view. What kills a VMI programme is not a modest transfer method but inconsistent stock records, uncounted locations, goods receipted late and items that exist under two codes. The discipline described in our guide to inventory management in the supply chain is the prerequisite, not an optional refinement, and the broader principles of inventory management apply just as firmly when someone else is doing the replenishing.
Agree the format and the cadence explicitly, and agree what happens when a file does not arrive. A supplier replenishing against last week's figures because Monday's export failed silently will eventually ship the wrong thing, and the argument that follows will be about trust when it should have been about a monitoring alert. Where a platform such as ProcureWave already holds your item master, receipts and consumption in one place, that export becomes a routine job rather than a monthly scramble.
Who owns the stock and when title transfers
Ownership is the clause people skip and later regret. In standard VMI, title passes on delivery exactly as it would in a normal purchase: the supplier decides the quantity, but once the goods are received they are yours, and they sit on your balance sheet. Under consignment, title stays with the supplier until a triggering event, usually consumption, issue to production, or sale to your own customer.
Whichever applies, the agreement must be unambiguous about four points: the precise event that transfers title, how consumption is reported and how often, who carries the risk of loss or damage while the stock is on your premises but not yet owned, and how counts are reconciled when your records and the supplier's disagree. Insurance follows risk, not title, so state it plainly rather than assuming.
The reconciliation clause deserves particular attention. Under consignment the supplier invoices against your reported consumption, so a stock count discrepancy is also an invoicing discrepancy. Agree the counting frequency, who attends and how differences are settled before the first delivery.
Benefits for both sides, and the real risks
VMI is unusual among procurement arrangements in that both parties can genuinely gain, which is why it survives. The buyer gains availability and time; the supplier gains visibility and stability.
- Fewer stockouts. Replenishment is triggered by actual consumption against agreed floors rather than by someone noticing a shelf is empty.
- Less buying admin. A category that generated dozens of small orders a month collapses into a single agreement and a review cycle.
- Better supplier forecasting. Real consumption data lets the supplier plan production and smooth their own capacity, which usually shows up in your pricing.
- Lower total inventory. Visibility on both sides removes the duplicate safety stock that each party holds against the other's uncertainty.
- Fewer expedites. Emergency shipments, overtime picking and premium freight all fall away when replenishment is planned rather than reactive.
The risks are equally real. Over-reliance is the most serious: hand a category entirely to one supplier and your ability to switch, or even to price-check, erodes quietly. Complacency is next, because once nobody internally is watching stock, nobody notices levels drifting upwards. Data quality problems compound rather than surface, since the supplier acts on bad figures faster than a human buyer would have. And obsolete stock creates the classic argument: the supplier says they replenished to the agreed maximum, you say nobody needed it, and the agreement is silent on who eats the write-off. Keep a credible alternative supplier qualified, keep reviewing, and settle the obsolescence question in writing up front.
What belongs in a VMI agreement, and which categories fit
A VMI agreement sits alongside your normal commercial terms and covers the specifics of the arrangement. At minimum it should state the item list with agreed minimum and maximum levels and the process for changing them; the data to be shared, its format and its frequency; the supplier's replenishment authority and any approval threshold; delivery and receipting arrangements; the ownership and title transfer position; the treatment of obsolete, damaged and slow-moving stock; the KPIs and review cadence; and the exit terms, including what happens to stock on site when the arrangement ends.
Exit terms are the clause most often missing and the one you will most want. Agree the notice period and, critically, whether the supplier buys back unconsumed stock and at what value. Without it, ending a VMI arrangement means inheriting a warehouse full of someone else's replenishment decisions.
On category fit, the pattern is consistent. VMI suits high volume, steady demand, relatively low value items from a small number of capable suppliers: fasteners, packaging, consumables, safety equipment, maintenance spares, basic raw materials. It suits poorly anything volatile, seasonal, heavily customised, short shelf life or strategically sensitive. If you cannot forecast an item within reason, handing it to a supplier does not make it forecastable; it just moves the guessing. This is also where your vendor management segmentation earns its keep, because VMI is only sensible with a supplier you have already assessed as reliable.
KPIs to govern a VMI arrangement
Because you have delegated a decision, you need measures that tell you whether it is being made well. Keep the set small and review it on a fixed rhythm with the supplier.
Service level
The percentage of demand met from stock. The primary test: if availability has not improved, the arrangement is not working.
Inventory turns
How fast stock cycles. Rising service with falling turns means the supplier is buying performance with your shelf space.
Level compliance
How often stock sits within the agreed minimum and maximum. Frequent breaches signal wrong levels or unreliable data.
Data accuracy
Counted stock against reported stock. The leading indicator for every other number on this list.
Add obsolete and slow-moving value as a standing agenda item, and treat trends rather than single readings as the signal. These reviews work best inside the wider performance rhythm described in our supplier relationship management guide, since VMI is precisely the kind of arrangement that only strategic suppliers should be trusted with.
How to set up VMI, step by step
Start narrow. Pick one supplier and one category that meets the fit criteria, and treat the first six months as a pilot with an explicit review date. The sequence that works is: clean your data first, then choose the category and supplier, then agree levels item by item using real consumption history rather than opinion, then decide the ownership model and write the agreement, then set up and test the data feed, then run in parallel with your existing ordering for a few cycles before handing over.
That parallel run is worth the effort. It shows you what the supplier would have ordered against what you actually ordered, and the differences are the most instructive conversation of the whole project. Only once the two lines track sensibly should you switch off the manual route. After that, review the levels quarterly, because the numbers you agreed at the start reflect a demand pattern that will have moved.
Extend deliberately rather than enthusiastically. A second category, then a second supplier, each with its own agreement and its own levels. VMI is one technique within a much larger supply chain management toolkit, and it earns its place on specific categories rather than as a general policy. If you want to see how the underlying stock, consumption and supplier records can sit in one place so a VMI feed is a routine export rather than a project, get in touch and we will walk you through how teams run it on ProcureWave.
Frequently asked questions
What is vendor managed inventory?
Vendor managed inventory, or VMI, is an arrangement in which the supplier takes responsibility for monitoring your stock of their products and replenishing it to agreed minimum and maximum levels. Instead of you raising purchase orders when stock runs low, the supplier sees your consumption data and ships what is needed, within limits you have both signed up to.
Is vendor managed inventory the same as consignment stock?
No. VMI is about who decides when to replenish; consignment is about who owns the stock while it sits on your shelf. You can run VMI where you buy the goods on delivery, and you can hold consignment stock that you still order yourself. The two are often combined, but they are separate decisions and should be written up separately in the agreement.
What data does the supplier need for VMI to work?
At minimum, current on-hand quantity by item and location, consumption or sales over a recent period, open orders already in transit, and any known changes in demand. Accuracy matters more than frequency, because a supplier replenishing against wrong stock figures will make wrong decisions faster than you ever could. Our inventory management guide covers the record-keeping that makes this feasible.
Which product categories suit VMI best?
High volume, steady demand, low value per unit and a small number of suppliers. Fasteners, packaging, consumables, maintenance spares, cleaning products and basic raw materials are classic candidates. Volatile, seasonal, highly customised or short shelf life items are usually poor fits because the supplier cannot forecast them any better than you can.
What are the main risks of vendor managed inventory?
Over-reliance on a single supplier, internal complacency once nobody is watching stock, poor data quality feeding bad replenishment decisions, and disputes over slow-moving or obsolete stock that the supplier pushed in. All four are manageable, but only if the agreement sets limits and someone on your side still reviews performance.
Want to see this in your own numbers?
Book a tailored demo and we will show ProcureWave running on scenarios that match your business.
Get in touch
