The Reorder Point Formula Behind an Amazon FBA Side Business

Sat, 03 Oct 2026 15:42:43 +0000

A lot of developers here have thought about selling a physical product on the side, and the part that decides whether it works looks a lot like capacity planning. Here is the inventory math Amazon FBA sellers run, written the way we would write it for a system.

The Inputs: Velocity and Lead Time

Daily sales velocity comes from your 30, 60 and 90 day averages. The 30 day number drives short term reorders, the 90 day number drives longer planning, and if sales are trending up you weight the 30 day number more heavily.

Lead time is the stack of everything between placing an order and units going on sale. For a private label product from China that is production (15 to 30 days), sea freight (25 to 35 days), customs (2 to 5), Amazon check in (5 to 14) and a delay buffer (7 to 10), often 55 to 95 days in total. Domestic wholesale is much shorter, about 12 to 27 days.

The Formula

Reorder point = daily velocity x (lead time + safety stock days). At 15 units a day, a 75 day lead time and a 10 day buffer, that is 15 x (75 + 10) = 1,275 units. When stock at Amazon drops to 1,275, the next order goes in.

Reorder quantity targets 60 to 90 days of coverage, so 75 days at 15 a day is 1,125 units. Then adjust for supplier minimums and price breaks, since 1,500 units at $4.20 can beat 1,000 at $4.80. The Amazon FBA inventory management guide walks through each step.

The Two Failure Modes

A stockout makes the listing inactive, pauses ads, and starts eroding organic rank within 24 to 48 hours. A 7 day stockout can take 3 to 4 weeks to recover from.

Overstock fails slowly instead. Storage is $0.56 per cubic foot most of the year and $2.40 from October through December, and units older than 181 days add an aged inventory surcharge. Amazon also scores you on this with an IPI from 0 to 1000, and below 400 it limits how much you can store.

The Takeaway

Treat it like a monitored system. Recompute velocity monthly, check levels twice a week, and pad Amazon's own Restock tool, which tends to underestimate international lead times. The complete Amazon FBA guide covers sourcing, listings and fees around this math.

Beyond the Model: Why AWS Spent $8B on Generators & What It Means for Cloud Architecture

Fri, 18 Sep 2026 15:57:55 +0000

Hey Legions,

When we think about scaling AI, we usually talk about model params, CUDA kernels, or vector search efficiency. But there’s a physical limit we’re rapidly hitting: Raw Grid Power.

Amazon Web Services (AWS) just locked in an $8 Billion contract with Generac (plus a 2.6% stock warrant) to supply industrial backup generators for their AI data centers. First $2.4B in hardware lands in 2027-2028.

Here is a quick breakdown of the backend realities driving this deal:

Checkpoint Loss Risk: Training massive models takes weeks across thousands of GPUs. A 100ms grid flicker doesn't just drop a connection; it corrupts active memory states and destroys millions of dollars in compute time.

Continuous Heavy Load: A hyperscale AI data center draws 100 to 500 Megawatts continuously. That is equivalent to powering up to 375,000 homes.

Grid Upgrade Mandates: With legislation like the Ratepayer Protection Act passing, cloud providers are now forced to pay full infrastructure costs to upgrade public grids.

The Hardware Bottleneck: Power availability—not just GPU allocation—is now the primary bottleneck for scaling LLMs and cloud clusters.

As devs, are you seeing infrastructure costs or cloud provider power limits affecting how you architect your backends, or are you mostly relying on managed APIs?

👉 Read my full deep-dive analysis here: [https://www.thefluxread.com/2026/09/amazon-just-spent-8-billion-on.html]