Sourcing

Knowing What to Order, and When

Knowing What to Order, and When

Besides SKU.io I run a company that resells other brands’ products on Amazon, eBay and Walmart. In April I was on our purchasing call, sharing my screen, looking at one of those brands. We hadn’t reordered it in months. The reorder forecast came back with a small order at 90 days of stock. “If we wanted to go crazy let’s just go 180,” I said, and it was still small. The supplier’s minimum is around $10,000, which at that pace was about a year of inventory. Cash was tight and the lowest projected stockout was 48 days away. So we waited.

On that same call I said I didn’t think the profit number on the sales report was right, and then went straight back to stock.

Projected stockout date and suggested order, both worked out from average daily sales

On September 2 I was scrolling the open purchase orders and thought the one for that brand was mislabeled. It wasn’t. I’d been reading 644 and it said 0664. “So this one I think we’re not gonna order because I don’t think it makes money,” I said. “I’ve done the calculations and I can’t see how it’s, like, profitable.”

Then I opened the ads for the same brand. The brand had agreed to cover up to $6,000 of spend as an experiment, and I’d been letting an AI tool run the campaigns. I saw its update every day. “I’ve been managing it with AI, so I don’t really know,” I said, and pulled up the dashboard. Ad spend as a share of the sales it brought in read 71%, 141%, 51%. One campaign was at 212%. We owed the brand a report on September 15. I switched the chart to revenue only to see if the ads had at least moved sales. “It actually went down in August.”

Then I asked whether the Walmart shipment had arrived.

Our accounting shows a thin profit on that brand for July through September, about $3.80 a unit, around 16%. That’s on stock we bought at old costs. My September math was for the next order, at today’s costs and today’s ad rates, and that’s the one I couldn’t make work. The forecast never had a field for either. In April it said wait because sales were slow.

We’re still talking to the supplier about some other arrangement.

SKU.io is what showed the stockout dates and the per-channel profit alerts in this post. If you sell on more than one channel, book a demo. If you sell on one, a spreadsheet or inFlow will do the forecast fine.

Newsletter

Get the next one by email

Short, practical writing on inventory, fulfilment and multi-channel operations — sent when there is something worth sending.

Double opt-in — we send a confirmation link and you are not on the list until you click it. One-click unsubscribe on every email. See our privacy policy.

Ready to see SKU.io in action?

Bring your real numbers and the messy questions — we'll walk the system against an operation like yours, not a slide deck.

Book a Demo

Founder-led demo · 2-day onboarding · no onboarding fees