Can't tell why your sales dropped? Sledge reads your store data and ranks the leaks by what fixing them is worth. Free for 14 days →

// Level 05 Making It Last · 5.1

Illustrated distant city in violet framed by a vine arch, with a golden moon in the one gap.

Metrics that matter

Updated August 2026 7 min read
SAME REVENUE. ONE OF THEM IS DYING. STORE A STORE B Revenue this month $12,000 $12,000 Conversion rate 2.4% 1.1% Sessions 16,600 4,400 Average order value $30 $248 Cost to get the order $14 $62 CONTRIBUTION PER ORDER −$0.50 +$74.40 ILLUSTRATIVE, AT 45% MARGIN FOR A AND 55% FOR B. THE TOP FOUR ROWS ARE ON EVERY DASHBOARD.

Illustrative figures. The pattern is not: a higher conversion rate on a cheaper product at a thinner margin routinely loses to the opposite.

The short answer

Conversion rate is the most quoted number in ecommerce and it decides almost nothing. It is a ratio with no money in it, so a store can raise it by discounting into a loss and the chart will applaud. The four numbers that actually decide whether a dropshipping store survives are contribution per order, blended acquisition cost, repeat rate and cash conversion. Only one of those appears on a standard dashboard, and it is not the one most people watch.

The problem with a ratio

Conversion rate is orders divided by sessions. That expression contains no price, no margin, no cost, and no currency at all.

Which means it can be improved by things that make you poorer. Cut prices by a third and conversion rises. Offer a discount code sitewide and conversion rises. Buy cheaper, lower-intent traffic and conversion falls even though nothing about the store changed.

Chapter 2.1 made this point in passing and it deserves the full version here: a 0.9% conversion rate on a $280 product beats a 2% rate on a $60 one, and no dashboard sorted by conversion rate will ever tell you that.

Conversion rate is a ratio with no money in it. You can improve it all the way to insolvency. The Dropshipping Playbook

The four that decide it

MetricWhat it answersWhere it comes from
Contribution per orderDoes a sale make moneyAssembled by hand. Not in the admin
Blended acquisition costIs the treadmill slowing downAd spend over total orders
Repeat rateAre you building an assetCustomers with 2+ orders in a fixed window
Cash conversionCan you pay for next weekPayout timing against spend timing

Contribution per order

The one number that makes every other decision in Level 04 legible. Selling price, minus product and shipping cost, minus fees, minus what the order cost to acquire.

It is the reason bundles can lose money while average order value rises, per chapter 4.2, and the reason a campaign discount costs more than its headline percentage, per chapter 4.5. If you track one number that is not already in your admin, make it this one.

Blended acquisition cost

Total acquisition spend divided by total orders, including the orders you did not pay for. This is where retention shows up as money rather than as a percentage: chapter 4.6 works through why moving three orders in ten to returning customers cuts it by about a third.

Watch the trend rather than the level. A blended CPA that falls month on month means the store is accumulating something. One that is flat means you are renting your revenue.

Repeat rate

Customers with more than one order, over total customers, in a fixed window. Fix the window or the number drifts upward as the store ages and tells you nothing.

Reported ecommerce averages sit near 28%, with an enormous spread by category, so compare against your own store last quarter rather than against a cross-category figure.

Cash conversion

The one that is specific to this business model and is missing from every generic metrics list.

Chapter 1.3 documented stores with 20% of every transaction held for 90 to 120 days while ad spend and supplier invoices continued daily. Their profit and loss looked fine while the bank account emptied. If you run on a reserve, or on any delayed payout, track the gap between when revenue is earned and when it is spendable, because that gap is what actually kills stores in a growth month.

The metrics to demote

Not useless, just not decisive, and dangerous when they are the ones on the wall.

Revenue
Contains no cost. Rises reliably while a store discounts itself to death, which is exactly the failure in chapter 4.2.
ROAS
Ignores product cost, fees and refunds. 3x on a 25% margin product is a loss, per chapter 2.2.
Conversion rate
Good for comparing a store against itself. Meaningless across price points, and unreadable at all under 500 sessions.
Sessions
A denominator, and an inflatable one. Bot traffic raises it and drags every rate below it down, per chapter 3.4.
Followers, list size
Inputs, not outcomes. Useful only alongside what share of them buy anything.

One replacement worth adopting

If conversion rate is the habit, revenue per visitor is the better habit and costs nothing to switch to. It is revenue divided by sessions, so it contains both halves of the thing you care about, and it cannot be improved by discounting your way to more orders at lower value.

Better still, contribution per visitor, which is the same idea with costs subtracted. That is the number that answers "is more traffic worth buying", which is the only question the traffic decision actually turns on.

Under 500 sessions, none of this is readable

Chapter 2.1 put the reliability threshold at 500 sessions over 30 days. Below it, every metric on this page moves a full point on one good day. Track them anyway, so the history exists, but do not make decisions from them yet.

What a small store's dashboard should be

Not a dashboard. A short list, updated weekly, in whatever you already use.

  1. Contribution per order, and its trend over four weeks.
  2. Blended acquisition cost, and its trend.
  3. Repeat rate, in a fixed window, monthly.
  4. Cash available against next month's committed spend.
  5. One discovery number: zero-result search rate, or share of catalogue with no views, per chapter 4.9.
  6. What you changed this week, dated. That last line is chapter 5.2, and it is what makes the five above interpretable at all.

Common questions

So conversion rate is useless?
No, it is diagnostic rather than decisive. It is the right number for answering "did this store change get better or worse", where price and product are held constant. It is the wrong number for answering "is this business working", because it contains no revenue, no cost and no margin.
What is contribution per order exactly?
Selling price minus product cost, minus shipping, minus payment and app fees, minus the acquisition cost of that order. What is left contributes to your fixed costs and your time. It is the number chapter 2.2's break-even ROAS is really about, expressed per order instead of as a ratio.
Why does cash conversion matter if I'm profitable?
Because profitable stores run out of cash. Chapter 1.3 documented payment reserves holding 20% of every transaction for 90 to 120 days while ad spend and supplier payments continued daily. A store can be profitable on paper and unable to buy inventory in the same week.
How often should I look at these?
Weekly for the trend, monthly for the decision. Daily checking on a store under 500 sessions a month is reading noise, per chapter 2.1, and it produces the reactive changes that make attribution impossible, per chapter 5.2.
Illustrated violet ridges receding to a yellow sun, over a field of pale flowers.

One number pays for the rest

What a single order is worth decides what every other fix can afford. Chapter 4.1 is where that number moves.

Go to chapter 4.1