Inventory Alerts with AI agents

Neotask watches stock levels across your storefront and warehouse feed and tells the right person before a SKU actually runs out — not after a customer order fails. It polls Shopify inventory levels against reorder thresholds set per SKU, cross-checks incoming purchase orders so it doesn't cry wolf on stock that's already inbound, and posts a prioritized restock list to Slack ranked by how many days of runway remain. Ops teams stop discovering stockouts from customer complaints.

How it works today vs. with Neotask

Most inventory alerting is either too noisy or too late: a flat 'alert below 10 units' rule fires constantly for slow SKUs and misses fast-moving ones that can sell through 200 units in a day. Manually reconciling stock levels against pending purchase orders in a spreadsheet is the usual workaround, and it rots the moment someone forgets to update it after a shipment lands. Neotask fixes both problems by computing runway — units on hand divided by recent sell-through velocity — instead of a static threshold, and by pulling live PO status so a SKU that's already been reordered doesn't clutter the alert queue. The result is a restock list ranked by actual urgency, refreshed automatically as sales and shipments happen.

The agent flow

Pull live inventory levels

Neotask polls Shopify's inventory API on a schedule (hourly by default) for every active SKU, capturing units on hand per location.

Integration: shopify

Compute sell-through velocity

It calculates a rolling 14-day sales rate per SKU from order history, so a slow mover and a fast mover are judged by days-of-runway, not raw unit count.

Cross-check pending purchase orders

SKUs with an open PO already covering the shortfall are suppressed from the alert list unless the PO is now itself overdue against its expected arrival date.

Integration: quickbooks

Rank and post the restock list

Remaining at-risk SKUs are ranked by days-of-runway ascending and posted to the ops Slack channel with current stock, velocity, and suggested reorder quantity.

Integration: slack

Log the alert history to a tracker

Every alert and its resolution (reordered, discontinued, backorder accepted) is logged to a shared Airtable base so seasonal patterns are visible over time.

Integration: airtable

Variations

• Multi-warehouse operations run the same velocity math per-location instead of aggregated, since a SKU can be critical in one warehouse and fine in another.
• Manufacturers add a raw-materials layer, alerting on component stock that would block production of a finished SKU even if the finished SKU itself still shows units on hand.
• Marketplace sellers extend the Shopify pull to also cross-reference Amazon or other channel inventory feeds so a single restock decision accounts for total demand across channels.

Frequently asked questions

How is this different from Shopify's built-in low-stock notifications?

Shopify's native alerts use a flat threshold per SKU with no sell-through context and no PO awareness, so they either fire too early on slow SKUs or too late on fast ones. Neotask's runway calculation and PO suppression solve both.

What happens with a brand-new SKU with no sales history?

It falls back to a manual threshold you set at launch until 14 days of order data accumulate, then switches to velocity-based runway automatically.

Can it also flag overstock, not just understock?

Yes — the same velocity model flags SKUs with more than a set number of months of inventory on hand, useful for cash-flow and clearance planning.

Does it place reorders automatically?

By default it only alerts and recommends a quantity; auto-placing POs is available as an opt-in step for SKUs you mark as pre-approved for automatic reorder.

Plans

Continue