Stock management is the practice of knowing what you have, deciding what to reorder and when, and making sure the number in your system matches the number on the shelf.
That last part is where most businesses lose money without noticing. Research from Auburn University’s RFID Lab puts average US retailer inventory accuracy near 63%, which means roughly a third of inventory records don’t match reality. If you also deliver what you sell, that gap shows up as a driver standing in a cooler at 6 a.m. looking for a case that was sold twice.
These are the stock management habits that hold up when your goods leave the building on your own vehicles: setting par levels, calculating reorder points, sizing a buffer, counting often enough to catch drift, and knowing which items deserve your attention.
Metrobi drivers are rated 4.97/5
Trusted by local businesses for:
- Background-checked professionals
- Specialized in business deliveries
- Same drivers for consistency
- 4.97/5 average delivery rating
The Bottom Line
- Accuracy comes before software. A stock management system built on numbers nobody trusts just reports the wrong figure faster.
- Par levels are the floor, reorder points are the trigger. Reorder point = average daily usage × supplier lead time, plus safety stock. Without lead time in the formula, every reorder is already late.
- Count a little, often. Cycle counting a handful of items weekly finds errors while they’re still small. The annual stocktake finds them in January.
- Shrink is bigger than people assume. The National Retail Federation’s 2024 survey put average shrink at 1.68% of revenue, about $112 billion across US retail.
- Committed stock isn’t available stock. Anything loaded for tomorrow’s route is already spoken for, and selling it twice is the most common error in a delivery operation.
Stock Management and Inventory Management Are Not the Same Thing
The terms get used interchangeably, and mostly that’s fine. There is a real difference worth knowing.
Inventory management covers everything you hold: raw materials, packaging, work in progress, finished goods, the spare parts for the van. Stock management is the narrower job of controlling the finished goods you sell: quantity, condition, and location.
For a bakery, flour is inventory and the boxed dozen is stock. For a wholesaler, they’re nearly the same thing. The distinction matters when you buy software, because some systems only handle finished goods and will not track an ingredient through production.
What both share is the same goal: hold enough to meet demand, without tying up cash or cooler space in product that won’t move. Everything below is about hitting that balance on purpose rather than by feel.
What Poor Stock Control Actually Costs
Running out is the loss people notice. It isn’t the only one.
- Lost sales from stockouts. NielsenIQ has estimated out-of-stock issues cost the US retail industry around $48 billion a year. At your scale it shows up as a standing customer finding another supplier.
- Cash sitting in the wrong product. Overstocking converts working capital into boxes. IHL Group put the global cost of inventory distortion — overstocks and stockouts together — at $1.77 trillion.
- Shrinkage. The NRF’s 2024 National Retail Security Survey put average shrink at 1.68% of revenue, the highest in more than a decade. Theft accounted for close to two-thirds of it.
- Spoilage. For food, floral and anything refrigerated, time is the main destroyer of value, and it doesn’t wait for your count.
- Delivery failures. A short case discovered at load time means a partial delivery, an apology, and often a second trip, which is the single most expensive thing a delivery operation can do. The companion to this guide, supply chain management software for last-mile operations, covers the tool layer that connects stock numbers to the route so that failure stops happening.
Set Par Levels Before You Buy a Stock Management System
A par level is the minimum quantity you want on hand for an item before you reorder. Below par, you’re exposed. The discipline of setting them is worth more than any software feature.
Setting them is simple arithmetic plus judgment:
- Start with demand. Average weekly usage over a representative recent period. Use eight weeks if you can, and exclude the one bizarre week.
- Adjust for variability. An item that sells 40 a week every week needs a smaller cushion than one that swings between 10 and 90.
- Adjust for consequence. Running out of your signature product costs more than running out of a slow-moving line. Par levels should reflect that.
- Review quarterly, and before every season. A par level set in February is wrong by May in most food and floral businesses.
Write them down somewhere that isn’t one person’s memory. A par level that only exists in the head of the person who does the ordering is a staffing risk, not a system.
How to Calculate a Reorder Point
Par level tells you how low you’re willing to go. The reorder point tells you when to place the order so you never get there. The difference is supplier lead time, and ignoring it is the single most common stock management error.
Reorder point = (average daily usage × lead time in days) + safety stock
Work an example. You sell 30 units a day. Your supplier takes 4 days to deliver. You hold 40 units of safety stock.
Reorder point = (30 × 4) + 40 = 160 units
When stock hits 160, you order, not when it hits 40. By the time the delivery lands you’ll have burned through 120 units and be sitting on your buffer, which is exactly what the buffer is for.
Two things break this formula in practice. Lead time that varies more than you think, and usage that spikes on a day of the week. Both are handled by the safety stock term rather than by fiddling with the average.
Safety Stock: How Much Buffer Is Enough
Safety stock is what covers you when demand runs hot or a supplier runs late. Too little and you stock out. Too much and it’s cash in a cooler.
The practical formula for a small operation:
Safety stock = (maximum daily usage × maximum lead time) − (average daily usage × average lead time)
Using real worst cases from your own history, not hypotheticals. If your supplier’s worst delivery in the past year was 7 days instead of 4, 7 is the number that belongs in the calculation.
A few rules that keep buffers sane:
- Buffer the items that hurt, not everything. Safety stock on every SKU is just overstocking with a formula attached.
- Perishables get time-limited buffers. A week of safety stock is useless if the product dies in five days. For those items, the buffer is a second delivery day, not a bigger pile.
- Recalculate when a supplier changes. New supplier, new lead time, new buffer.
Cycle Counting Beats the Annual Stocktake
Counting everything once a year tells you how wrong you were, eleven months too late. Cycle counting, which means counting a small subset frequently on a rotation, finds the same errors while they’re still cheap to fix.
A workable cadence for a small operation:
- Weekly: your top 10 to 20 moving items. These are where errors compound fastest.
- Monthly: anything perishable or high-value, regardless of movement.
- Quarterly: the long tail.
- Always: after any incident — a big event order, a delivery dispute, a new hire’s first week.
Count at the same time of day, ideally before the first order cutoff, so you’re not chasing stock that’s mid-pick. And count the shelf first, then compare to the system. Looking at the expected number before you count is how a count becomes a confirmation of what you already believed.
When the count and the system disagree, the investigation matters more than the correction. Adjusting the number without asking why it drifted means you’ll adjust it again next month.
ABC Analysis: Which Stock Deserves the Attention
Not every item earns the same scrutiny. ABC analysis sorts stock by its share of annual value so you can spend your attention where the money is. A common split looks like this:
| Class | Roughly this share of SKUs | Roughly this share of value | Counting cadence | Buffer policy |
|---|---|---|---|---|
| A | A small top slice | The large majority | Weekly | Tight par, deliberate safety stock |
| B | The middle band | Moderate | Monthly | Standard par, modest buffer |
| C | The long tail | A small fraction | Quarterly | Generous buffer, order in bulk |
The exact percentage thresholds matter less than the behavior they produce. A-class items get counted often and bought carefully. C-class items get bought in larger quantities less often, because the carrying cost is lower than the hassle of managing them tightly.
Run the sort once a quarter. Items migrate between classes, usually right after you introduce or drop a product line.
Stock Management for Perishable Goods
If your product has a date on it, standard stock logic needs one addition: the oldest stock has to move first, every time.
First-expiring-first-out is the rule, and it’s enforced physically, not in software. New deliveries go behind existing stock on the shelf, not in front of it, because the path of least resistance for a person picking an order in a hurry is to grab the nearest box.
Practical habits that hold up under pressure:
- Date-label on receipt, not later. A case without a visible date will be picked at random.
- Zone by date, not just by product. A “use this week” shelf does more than a label on each box.
- Count perishables more often than value alone suggests. Spoilage doesn’t show up in a sales report until you look for it.
- Track waste as a line item. If spoiled product is written off without a record, nobody learns which par level was too generous.
Keep Committed Stock Separate from Available Stock
This is the rule that distinguishes stock management in a delivery business from stock management in a shop.
When a customer buys off a shelf, availability and ownership change at the same moment. When you take an order on Monday for delivery Wednesday, the stock is spoken for on Monday but still physically present until Wednesday morning. Anyone looking at the shelf, or at an on-hand number, sees product that isn’t available.
So the number that matters is available-to-promise: on hand, minus everything committed to orders not yet delivered.
- Reserve stock at order entry, not at pick time. By pick time it’s too late to tell the customer.
- Stage tomorrow’s orders physically. A separate area, cage or shelf for allocated stock prevents the double-sell more reliably than any field in a system.
- Reconcile after loading, not before. What actually went on the van is the number that counts.
Getting this right is largely a systems question once your volume grows past what one person can hold in their head. The guide to supply chain management software for last-mile operations covers which tools handle available-to-promise properly and how the stock layer connects to order capture and routing.
Stock Management KPIs Worth Tracking
Four numbers tell you whether your stock control is improving. More than four and nobody looks at any of them.
- Inventory accuracy. Counted units divided by system units, per cycle count. This is the health metric; everything else is downstream of it.
- Inventory turnover. Cost of goods sold divided by average inventory value. Rising turnover usually means tighter control. Falling turnover means cash is accumulating in stock.
- Stockout rate. The share of order lines you couldn’t fill in full. Track it by item, because the fix is usually one or two SKUs.
- Shrink and waste as a share of revenue. Compare against the NRF’s 1.68% retail average to see whether yours is normal or a problem.
Review them monthly. Weekly is noise, and quarterly is too late to act on a trend.
Common Stock Management Mistakes
- Ordering by feel on a busy morning. The reorder point exists precisely so this decision isn’t made under pressure.
- Treating on-hand as available. The double-sell that follows is the most expensive error on this list.
- Counting only what’s easy to count. The items in the back corner are where discrepancies live.
- Writing off spoilage without recording it. Waste you don’t measure becomes a par level you never fix.
- One person owning all of it. When the person who knows the stock takes a week off, an undocumented system becomes visible fast.
- Buying a stock management system to fix a discipline problem. Software enforces a process. It can’t invent one.
Frequently Asked Questions
How often should I count stock? Rotate rather than counting everything at once. Your fastest-moving and highest-value items weekly, perishables monthly at minimum, the long tail quarterly. The goal is to touch every item several times a year rather than once.
What is a good inventory accuracy rate? Aim above 95% on your A-class items. The ~63% average that Auburn’s RFID Lab research found across US retailers is a warning rather than a target.
Does stock management software replace counting? No. It tells you what the count should be and flags when something drifts. Physically verifying is still the only way to know what’s actually on the shelf.
How do I handle seasonal demand? Recalculate par levels and reorder points before the season starts, using the same period from last year as your baseline. Also confirm supplier lead times, since most get longer at peak, which pushes reorder points up before your demand does.
Is a spreadsheet good enough for stock management? For a small, stable range it can be, as long as it’s updated on a fixed schedule and more than one person knows how it works. It stops being enough when you need to reserve stock for future delivery dates, because that’s where manual tracking breaks first.
Sources
- Auburn University RFID Lab inventory accuracy research, as reported by CYBRA
- National Retail Federation, National Retail Security Survey
- IHL Group, Fixing Inventory Distortion
