The patient is in the chair. The doctor reaches for the supplies. There's nothing in the drawer. Someone is dispatched to the storage room. Three minutes later, also nothing. Five minutes later, a junior staff member is on Amazon ordering same-day delivery for an item that was supposed to be ordered last Wednesday.
Stockouts in clinics are nearly always organizational, not procurement, failures. Here's how to set up inventory so this conversation stops happening.
Table of Contents
- Why clinic inventory goes wrong
- The four core mechanisms
- Par-level math: how much to keep on hand
- Cycle counting beats annual count
- Vendor management without spreadsheets
- What to measure
- FAQ
Why clinic inventory goes wrong
- No one owns it formally; everyone assumes someone else is watching.
- The "system" is a spreadsheet someone updates monthly.
- Reorder is reactive ("we're almost out") rather than proactive (par-level triggers).
- No visibility into expiration dates — items expire before being used.
- Receiving isn't matched to invoices; theft and shrinkage go undetected.
The four core mechanisms
| Mechanism | What it does |
|---|---|
| Item catalog with par levels | Each SKU has a min, max, and reorder point |
| Usage tracking | Items consumed are logged (manually or via integration with visit types) |
| Automated reorder triggers | System flags or auto-orders when below threshold |
| Cycle counting | Small portion counted each week, not everything once a year |
Par-level math: how much to keep on hand
Par level = (average daily usage × lead time in days) + safety stock. Safety stock is typically 1-2 weeks of usage for routine items, more for critical ones.
Worked example: gloves used 12 boxes/week. Vendor delivers in 4 days. Safety stock 1 week.
- Average daily usage: ~1.7 boxes
- Lead time stock: 1.7 × 4 = 6.8 ≈ 7 boxes
- Safety stock: 12 boxes
- Reorder point: ~19 boxes
- Max (par): 30 boxes
When inventory hits 19, the system flags or auto-orders. You never run out, you don't over-stock.
Cycle counting beats annual count
The traditional approach: shut down for half a day once a year, count everything, write off the discrepancy. This is the worst possible model — the variance has been growing for 12 months and nobody knows when it started.
Cycle counting: every week, count one section of the inventory. Twelve weeks, full inventory covered. Variances surface within days, not months. The annual count becomes a verification, not a fire drill.
Vendor management without spreadsheets
- One vendor catalog per item, with backup vendor flagged.
- Receiving matched to PO at the moment goods arrive.
- Invoice reconciliation captured at receiving, not at month-end.
- Lead time tracked per vendor (not advertised — actual).
- Performance metrics: on-time delivery, accuracy, price stability.
The clinics with the smoothest supply have 1-2 primary vendors and aren't shopping around month to month for $5 savings on each item. Time costs more than the discount.
What to measure
- Stockout incidents per month.
- Inventory turn rate (how many times the inventory cycles per year).
- Expired items written off.
- Variance at cycle count (target: under 2%).
- Carrying cost as % of monthly revenue.
FAQ
Does my clinic management platform need a built-in inventory module?
For a single-clinic, yes — separate inventory software adds login overhead. For multi-branch with central warehousing, a dedicated tool may be needed.
How do I handle expiration dates?
Track them at receiving. The system surfaces items expiring within 30 days for use-up; 7 days becomes a hard alert.
What's a healthy inventory turn rate?
For most clinics, 6-12 turns per year. Below 6 means too much capital tied up; above 12 risks stockouts on demand spikes.
Should I track every SKU, including pens?
No. Use ABC analysis: A items (top 20% by value, ~80% of cost) get rigorous tracking. C items (low value) get simpler "two-bin" approaches.
How do I prevent theft / shrinkage?
Cycle counts catch it early. Role-based access on inventory write-offs catches the rest. Most clinic shrinkage isn't theft — it's expiration and process loss.
Can the system auto-order without owner approval?
Yes, but pick your threshold. Low-cost A-list items can auto-order; expensive items should need a human approval. Mostly hands-off, deliberately not all of it.
The bottom line
Inventory is the most boring problem in clinic management — until it interrupts a visit. The fix is not heroism. It's a small set of mechanisms (par levels, cycle counts, vendor management) running on software that does the math. Set it up once; reap the calm forever. Pair this with our paperless clinic checklist for the broader operational layer.