Blog
Commerce data analytics, technology and news.
At what percentage should an alert fire? Thresholds calculated from a year of sales at two brands
We calculated alert thresholds from a year of daily sales at two brands. The commonly used ±15% fires on half of all day

Build 812 charts and 79% of your metrics still aren't on screen — the questions dashboards don't cover
We took apart the custom dashboards of six brands. The number of charts they had built ranged from 11 to 812, a 73x diff

We Run 16 Channels, but 95% of Revenue Comes from Three — The Cost of Multichannel Aggregation
We took apart the channel mix of six brands. The number of channels they operate ranges from 4 to 33 — a 9x spread — yet

When “Spend More on What’s Working” Is the Most Expensive Decision — Growth Headroom Through Marginal ROAS
Budget increases should be judged on marginal ROAS (the revenue generated by the next won you spend), not average ROAS.

The Price Tag on “It’ll Sell Eventually”: How to Identify and Clear Deadstock
The most expensive mistake is believing that leaving stock alone costs nothing. When we broke down this brand’s inventor

Our No. 1 Revenue Channel Had the Lowest Margin — How to Calculate Net Margin by Channel
We ranked the same channels twice — once by revenue, once by net margin rate — and the rankings came out almost perfectl

Settlement Says 72.5, the Bank Keeps 21.1 — Breaking Down Fulfillment Channel P&L
If GMV is 100, the number on your settlement dashboard is 72.5. What you actually keep is 21.1. The 51.4 in between isn'

Subtract COGS and Your No. 1 Product Changes — But Not If You Stop at Gross Profit
Ranking products by revenue and by gross profit produced what was essentially the same list (rank correlation 0.99). The

The Economics of the First Repurchase — “We Want a Higher Repurchase Rate.” But Where Exactly Do You Push?
If your 1st→2nd purchase conversion is 45% and your 2nd→3rd is 72%, the problem isn’t your “repurchase rate” as a whole

Why “How much did we sell today?” has two answers — the gap between same-day sales and confirmed sales
If orders come in at 100, only 91.03 ends up confirmed. The real problem isn't the size of the gap — it's the timing. Th
