Last updated: July 31, 2026
NetSuite shows real-time inventory the moment you finish configuring item records with reorder points, lead times, and per-location stock levels. Most companies never finish that part, so they run a live system on last Tuesday’s numbers. The data is already in there. Nobody wired it to a decision.
A sales rep at a wholesale client sold the same pallet twice in one afternoon.
Not out of malice. He was looking at a number in NetSuite that said 340 units, and the number was correct, and it was also useless, because 260 of those units were already committed to an order that hadn’t shipped yet. He didn’t know the difference between On Hand and Available. Nobody had told him. Why would they, when the column he was trained to look at just says quantity and nothing about that word suggests there are five other quantities sitting one field away?
That’s the whole problem in one story. The system knew. The person didn’t.
Since this is a staffing company’s website, here’s who I am and what I sell. ERP and business systems consulting, which means I profit when you decide your inventory setup is a mess. Fair warning. I’ll also point out which parts of this you can knock out yourself on a slow Thursday, because charging for those would be embarrassing.
This is the inventory half of a larger argument I made about NetSuite optimization after go-live. Same thesis, narrower target. You already own the tools. Nobody switched them on.
What Real-Time Inventory Actually Means Here
Real-time inventory in NetSuite means every transaction touching stock, whether a sales order, a receipt, a fulfillment, or a transfer, updates item quantities the instant it posts. There is no overnight batch and no sync window. Accuracy depends entirely on whether your people record those transactions when they happen rather than later.
So the platform is real-time. Your data might not be.
That gap is where every stockout conversation actually lives. NetSuite isn’t withholding anything. It parrots back whatever the last person told it, whenever they got around to telling it, so if the warehouse picks a pallet Monday and records it Wednesday, you just spent two days with a lie in your ERP wearing the same confident little number in the same confident little column as the truth.
Nobody notices. Not until something ships short.
Which is why I get suspicious when a client tells me their inventory is “pretty accurate.” Pretty accurate is a feeling. Ask what the last cycle count variance was. Watch the room go quiet.
The Fields That Decide Whether You See It Coming
Inventory planning in NetSuite lives on the item record. Not in some separate module you have to go buy. On the record, in fields that are probably blank right now.
Go look. I’ll wait.
Here is what they do and who should own them. That right-hand column is the part almost everybody gets wrong, because these fields get assigned during implementation to whoever happened to have capacity that week rather than to the person who actually knows what the number should be, and then they stay assigned that way for years.
| Field | What it actually does | Who should own it |
|---|---|---|
| Reorder Point | The quantity that triggers “buy more.” Everything else on this list exists to make this number honest. | Buyer or demand planner |
| Safety Stock Level | Your buffer against a supplier being late or a week being weird. Set per location, not globally. | Ops, with finance sanity-checking the carrying cost |
| Lead Time | Days from PO to dock. NetSuite can calculate it from your actual receipt history instead of the number your vendor promised. | Purchasing |
| Preferred Stock Level | Where you want to land after replenishing. Drives suggested order quantity. | Buyer |
| Auto-Calculate flags | Lets NetSuite derive reorder point and lead time from demand history rather than a guess somebody typed in 2022. | Admin, once ops agrees on the lookback window |
| Item Location config | Per-warehouse versions of everything above. Skip this and a stockout in Reno hides behind surplus in Atlanta. | Admin plus each site lead |
That last row causes more damage than the other five combined.
Company-level quantities average away the thing you needed to know. You’re not out of stock nationally, congratulations, but the customer who ordered from the Reno warehouse is still getting a delay notice, and the surplus solving their problem is 1,900 miles away in a building with no transfer order pointed at it. Oracle documents lead time and safety stock per location as distinct settings for exactly this reason. Multi-location businesses that plan at the company level are running one number for several realities.
The auto-calculate flags are the cheapest win on the list. Your vendor’s stated lead time is a sales document. Your receipt history is evidence. Trust the evidence. Oracle’s Advanced Inventory Management documentation covers how demand-based replenishment pulls from that history, and switching it on is an afternoon of admin work rather than a project.
Your Quantity Available Is Lying to You
Back to the rep who sold the pallet twice.
NetSuite carries several quantity fields per item and they mean genuinely different things. People treat them as synonyms. They aren’t. The gap between two of them is where your next embarrassing customer call is currently incubating.
- Quantity On Hand. Physically in the building, per the system. Includes units already promised to someone else. This is the number everyone quotes and the number nobody should quote.
- Quantity Committed. Spoken for by open sales orders. Real inventory, already owned by a customer who doesn’t yet know they’re about to become part of a story.
- Quantity Available. On Hand minus Committed. The one your sales team should be reading.
- Quantity Back Ordered is demand you’ve already failed to meet. Treat it as an alarm, not a metric.
- Quantity On Order. Inbound on a purchase order. Useful for promising a future date, dangerous for promising today, and worth checking against the actual receipt history for that vendor before you say anything out loud to a customer with a contract.
- Then there’s In Transit, which matters enormously if you run transfers between sites and not at all if you don’t. Know which one you are.
Fixing this is not a systems project. It’s a screen layout and a conversation.
Take Available off the third tab and put it where the sales team already looks. Rename the field label in plain English if your people keep misreading it. Then tell them why it changed, and tell them what it cost the last time somebody read the wrong column, because that’s the step everyone skips and it’s the entire reason a perfectly good new field gets quietly ignored for the next six months.
Two hours of work. Maybe three.
Stockouts Are a Signal Problem Before They Are a Warehouse Problem
Here is my favorite uncomfortable statistic, and I want to be upfront that it is old.
Daniel Corsten and Thomas Gruen ran a worldwide retail audit and published the findings in Harvard Business Review in 2004, and they found an average out-of-stock rate of 8.3 percent. Roughly one shelf in twelve. What holds up better than the headline number is what shoppers did next: 31 percent bought the item somewhere else, 26 percent switched to a different brand, 19 percent took another product from the same brand, 15 percent delayed, and only 9 percent walked away with nothing. Their analysis of what stock-outs actually cost is more than twenty years old at this point and I still put it in front of executives.
Read that split again, because it’s worse than a lost sale.
Three quarters of those shoppers still bought something. Just not the thing you wanted them to buy, and in nearly a third of cases not from you at all. The revenue didn’t evaporate. It relocated. And your P&L logs that as a soft month rather than as a stockout, which is exactly why nobody in the building treats availability as a finance problem when it plainly is one.
Two decades later, with better systems than anyone had in 2004, the failure mode is unchanged. That should bother you more than it usually does.
The B2B version is quieter and more expensive. Nobody storms off. They just start dual-sourcing you, and eighteen months later your share of their spend has been cut in half through a series of small decisions nobody ever announced in a meeting.
You won’t see that coming either.
Nobody Wants to Talk About Cycle Counts
Every forecasting conversation I get pulled into eventually reveals itself to be a counting conversation.
Reorder points work off quantities. Quantities come from what the warehouse recorded. If your recorded quantity and your physical quantity have drifted apart, and they have, then every calculation stacked on top of that number inherits the error and hands it back to you with total confidence. Garbage in. Confident garbage out, at speed.
NetSuite has cycle counting built in. Bin management, count tasks, variance approval routing, all of it. Most mid-market accounts I open have never switched it on and instead shut the building for two days each December to count everything at once, which produces one accurate day per year and eleven months of decay.
Annual physical counts are theater for the auditors. Cycle counts are for you.
Start with the items that would actually hurt. Your top 50 SKUs by revenue, counted on a rotation, with someone owning the variance investigation and not just the recount. Because when variance shows up, a recount tells you the number. It won’t tell you why the number moved. The why is the only part that stops this happening again next quarter, and the why is the part that always gets skipped when everyone just wants the count to balance so they can go home.
What I’d Do in Your First Two Weeks
Not a transformation program. Four things, in order, and the order matters.
- Pull a list of items with a blank reorder point. Saved search, five minutes. On most accounts I audit the list is embarrassingly long, and roughly a third of it is items nobody has bought in two years, which is its own separate conversation about your item master.
- Check whether your locations are configured independently. If safety stock and lead time only exist at the company level, and you run more than one warehouse, stop reading this and go fix that first. Everything else on this list is worth less until it’s done.
- Ask the sales team which quantity field they read. Just ask them. The answer will tell you more about your inventory data than any report will, and I have never once had this conversation go the way the operations lead expected.
- Set up one alert. A saved search that emails your buyer when Available drops under the reorder point, scheduled daily. An hour of work. It replaces the current system, which is somebody noticing a problem while looking for something unrelated.
None of that requires a consultant. All of it requires an owner. Which brings me to the part I can’t fix with configuration.
Sometimes You Don’t Have a Data Problem, You Have an Empty Chair
I want to be honest about where software stops.
Reorder points drift because businesses change and nobody revisits the assumptions. Demand shifts, a supplier gets slower, you add a channel, and the numbers somebody configured during implementation quietly stop describing the company you’re running now. That’s not a NetSuite failure. It’s a nobody-owns-this failure, and I find it most reliably in companies that grew past $50 million in revenue without ever hiring the one person whose entire job would be keeping the system honest, because that headcount never looks urgent in the quarter you’re deciding about it.
What that person looks like depends on where it hurts.
Configuration drift and unowned item records point to an administrator. Planning logic that never matched how the business actually buys points to a functional consultant. And if your warehouse system and NetSuite disagree about what shipped, that’s an integration problem and a completely different skill set.
Hire the wrong one of those three and you lose a quarter.
KORE1 staffs all of them, which is why this piece is on their site rather than mine. NetSuite administrator staffing for the ownership gap. WMS integration consultants when the warehouse and the ERP are telling different stories. NetSuite integration specialists when the disagreement is broader than the warehouse, and supply chain IT staffing when the planning function itself is the hole. Their IT staffing services practice fills a typical search in about 17 days. If the work is a defined burst rather than a permanent seat, contract staffing is the cheaper shape and nobody should feel weird about it.
For manufacturers the calculus shifts again, since planning sits against production schedules rather than purchase orders. KORE1 covers that under manufacturing ERP staffing.
Questions I Get on Almost Every Warehouse Call
Can NetSuite actually predict a stockout, or does it just report one?
It predicts, provided you feed it. With reorder points, lead times, and safety stock configured per location, NetSuite flags items approaching their trigger before you hit zero and can suggest the replenishment order.
What it won’t do is invent a lead time you never recorded or account for a demand spike nothing in your history predicted. Prediction here means arithmetic on your own data, done continuously. Not clairvoyance. Anyone selling it as clairvoyance is selling something else.
We already run demand planning. Why are we still stocking out?
Usually the lookback window. Demand planning that averages twelve months of history will lag a business whose mix changed six months ago, and it will lag confidently.
Check the forecast method against your actual seasonality too. Moving average and seasonal average behave very differently on a product line with a real Q4, and I have watched a company run linear regression on a SKU with a hard holiday spike and then blame the software for the January pile of stock they were still sitting on in March.
How much of this is admin work versus something I need to pay for?
Most of it is admin work. Item record fields, saved searches, alerts, cycle count setup, and dashboard changes are all internal tasks if you have an admin with real hours protected.
Bring in outside help for integration architecture between NetSuite and a warehouse system, for multi-subsidiary planning structure, and for anything that changes how inventory hits the general ledger. Those spread. Dashboard cleanup doesn’t. Paying consultant rates for it wastes your money and bores mine.
Our warehouse team says the numbers are wrong. Are they?
Nine times out of ten they’re right that something is off, and wrong about what. Usually the physical count is fine and the timing is broken, with transactions recorded in batches hours after the work happened.
Run a cycle count on twenty high-velocity items and compare. Physical matches system? You have a latency problem, and the fix is process, maybe scanners. Physical doesn’t match? That’s a data integrity problem, and no forecast built on top of it means anything at all.
Do we need the WMS module, or is standard inventory enough?
Standard inventory handles a single building with straightforward picking. You need WMS or a third-party warehouse system once you have bins, wave picking, multiple pick paths, or labor you need to measure.
The tell is your people. If the warehouse runs on printed pick lists and tribal knowledge about where things live, you’ve outgrown standard inventory. And the integration between whatever you choose and NetSuite matters far more than which product you pick, which is the opposite of how most of these evaluations get run.
Is any of the AI stuff worth turning on for inventory?
Some of it, for the boring parts. Querying inventory in plain English instead of building another saved search is genuinely useful and available now through the NetSuite AI Connector.
What I wouldn’t do is hand demand forecasting to a model before your underlying counts are trustworthy. That just automates a wrong answer and makes it arrive faster. Fix the counting first. Then let the machine do the tedious lookups. That part it’s genuinely good at.
Go Look at Your Item Records
Your ERP already knows when you’re going to stock out.
The information is sitting in fields nobody filled in, on records nobody owns, feeding reports nobody scheduled. Not a software gap. Not a budget problem. It’s twelve hours of unglamorous configuration work that keeps losing to whatever is on fire this week, and it will keep losing until somebody’s name is attached to it.
Go pull the blank reorder point list. Right now, before this tab closes and the week eats you. Five minutes. It will tell you exactly how bad this is.
Bad list? Hit me up on LinkedIn and I’ll walk you through what I’d do about it. And if the honest answer is that you need somebody in the seat instead of another project, talk to the KORE1 team. That’s their lane, not mine.
