Stores rarely die of one dramatic mistake. They die of a number that was wrong for months while nobody looked at it. Here are the seven patterns behind most failures — and the specific metric that would have flagged each one early.
Survivorship content tells you what winners did. Failure analysis is more useful, because failures repeat: the same handful of numeric patterns, visible early, ignored until late. Each section below names the pattern and the number that exposes it.
The classic: revenue grows, the bank balance shrinks. Price minus product cost feels like margin, but shipping, packaging, payment fees, the tax slice, returns and ad cost per order eat it silently.
Early warning: per-order contribution margin — computed with every cost, updated when any cost changes. If you have never computed it, that is the afternoon project that saves the company.
Ecommerce pays suppliers today and collects from payment processors and marketplaces days or weeks later. Growth widens that gap: more orders = more cash out now against cash in later. Stores die profitable this way.
Early warning: a 4-week cash calendar — expected settlements in, supplier and ad payments out. The killer is not the loss, it is the surprise.
Bulk discounts push founders to buy deep; taste and seasons change; capital turns into boxes. Dead stock is a loss you have already taken but not yet admitted.
Early warning: weeks-of-stock per SKU and sell-through rate. Reorder winners in small fast cycles; discount losers early — the first markdown is the cheapest.
All revenue from one ad account, one marketplace ranking, or one algorithm is an unpriced risk. Accounts get restricted, rankings slip, CPMs double — none of it under your control.
Early warning: revenue share of the top channel. When one source passes roughly 70%, building the second leg (owned lists, another channel, another market) stops being optional.
Post-iOS pixels miss a large share of conversions; sellers over- or under-read ROAS and scale the wrong things. Winners get killed, losers get budget.
Early warning: ad-platform revenue vs actual order revenue, reconciled weekly. Fix the signal with the Conversions API before making budget decisions on it, and hold spend to your break-even ROAS.
When all growth is paid acquisition and none is repurchase, CAC inflation sets the expiry date. The auction gets more expensive every year; your customer list doesn't.
Early warning: repurchase rate and the LTV:CAC ratio. If second orders are rare, fix the post-purchase experience before scaling the ad budget.
Late shipments, unanswered inquiries, sloppy returns — each one small, together compounding into bad reviews, falling conversion and rising refund rates. Drift is invisible day-to-day and obvious in hindsight.
Early warning: time-to-ship, response time, refund rate — tracked as numbers, not vibes. All three are leading indicators; reviews are the lagging one.
None of this requires a data team. It requires the numbers to be in one place and looked at on a schedule — which is precisely the habit that separates the stores that survive year three from the ones that don't.
Numbers 1 and 2: per-order margin and the cash calendar. Everything else can be recovered from; selling at a loss with no cash cannot.
The fragmentation is itself the risk — metrics nobody consolidates are metrics nobody watches. Either build a weekly spreadsheet ritual or use a dashboard that aggregates store, ad and settlement data automatically; the point is one screen, on a schedule.
Sometimes — but a store with clean unit economics, cash visibility and a repurchase base gets many product attempts. The seven patterns above are what remove your right to keep trying.
DashBooster puts revenue, net profit, ad spend and settlement timing on one screen, every day — the early-warning panel this article describes.
Start your 7-day free trial →