Square Champion

Intelligent Min Stock Thresholds - Auto-Adjust Based on Sales Velocity

Screenshot 2026-04-19 112433.png

Title: Intelligent Min Stock Thresholds: Auto-Adjust Based on Sales Velocity

Category: Inventory Management


The Problem

Square lets merchants set a minimum stock alert for each item, but that number is static. Set it once, forget it, and it quickly becomes wrong.

A product that sold 2 units a week last winter might sell 8 a week in spring. Your min stock threshold doesn't know that. You either get alerted too late (stockout before you notice) or too early (alert fatigue from false alarms on slow movers).

 

For stores with hundreds of SKUs, manually reviewing and updating min stock thresholds is not realistic. Most merchants either never update them or abandon them entirely.

 

What We Actually Need

Min stock thresholds should be dynamic, automatically adjusted based on how each item is actually selling.

 

Specifically:

  • Velocity-based calculation: derive the recommended minimum from actual daily sales rate over a rolling 91-day window, not a guess at setup
  • Automatic daily adjustment: recalculate and update thresholds each morning as sales patterns change
  • New item protection: skip auto-adjustment for items without sufficient sales history so new arrivals aren't penalized with bad recommendations
  • Merchant override: let merchants lock a threshold on specific items to prevent auto-adjustment when they know something the system doesn't (seasonal buys, samples, display stock, promotional inventory)
  • Case pack awareness: round recommended minimums to case pack quantities so reorder suggestions align with how vendors actually ship
  • Max conflict detection: flag items where the recommended minimum meets or exceeds the current maximum and hold for merchant review instead of applying blindly
  • Skips slow and zero-velocity items: items with no meaningful sales history are excluded from adjustment automatically to avoid noise

 

Why This Matters

Stockouts are lost sales. Overstocking is tied-up cash. The gap between the two is min stock, and right now merchants are either guessing or building their own tooling to solve it.

I built daily auto-adjustment into my own inventory system because Square doesn't offer it natively. It runs every morning, adjusts thresholds based on 91-day rolling velocity, skips new items and pinned overrides, flags conflicts for review, and syncs the result back to Square. This morning it processed 111 items, adjusted 6, flagged 3 for human review, and correctly skipped 89 that were too new to have reliable data.

That shouldn't require custom development. It should be a native Square feature available to every seller.

 

The Ask

Intelligent, velocity-based min stock thresholds with daily auto-adjustment, merchant pin overrides, case pack rounding, and conflict detection built into Square's inventory management natively.

2 Replies
Square Champion

Interesting concept John @JTPets .   I see you are using a 91 day assessement.  In my business, it's a yearly cycle. 

 

For example, we go through "bowling shoes" a lot in August and December, and then it's slower the rest of the year.   I could see increasing the stock minimum in these months, but I would want it lower the rest of the year so I'm not sitting on stock (and in some cases, last years model).

 

Could the AI be smart enough to figure that out?

 

In related news...we started writing our own PWAs to address some industry specific needs that Square doesn't do...and the outlook is promising.

 

John Losito,
Sun Valley Lanes & Games

A flute without holes, is not a flute. A donut without a hole, is a danish.

Square Champion and Square Innovator
www.sunvalleylanes.com
Square Champion

The best software automation don't use ai to decide, they use ai to interpret ambiguous situations.  In this scenario there is no ai. Its rule based, every morning at 6am check our order generation logic. Which is also rules based. If an order would occur for an item and its days of stock would exceed 91 days before order, reduce the minimum. If we sold out increase it. But all of it can be adjusted based on the rules you set in your system.

 

This is why I use 3 months. It allows for a smooth rolling quarterly window of data for current trends. This all runs on a timer for me, but could be a manual trigger on demand if wanted or stopped in slower months. This is the joy of using ai to build rule based gates that an ai can later run or use for context and trigger it