Recovery ROI: which carts are worth a postcard (with a calculator)
Recovery mail has an unusually honest ROI question, because the unit is so concrete: should this one cart get a $1.50 card? Answer it with arithmetic once, encode the answer as a minimum-cart rule, and the program makes the right call automatically forever after.
The formula
Expected profit per card = cart value × gross margin × recovery rate − card cost.
Rearranged into the number you actually set: breakeven cart value = card cost ÷ (margin × recovery rate).
Example: $1.50 card, 60% margin, 5% postcard recovery rate → breakeven at $50. Every mailed cart above that has positive expected value; a comfortable rule doubles it and mails only carts above $100, where each card’s expected profit is clearly worth the operational attention.
Cart threshold calculator
Below breakeven, mailing that cart loses money even when it works. Set your minimum-cart rule at the comfortable line and every card that goes out has real expected profit behind it — that’s the arithmetic behind Return Address’s minimum-cart setting.
Keeping the inputs honest
- Card cost: use the all-in per-piece price ($1.50 here, printing and postage included). If your plan includes cards monthly, cards inside the allowance are effectively cheaper — a bonus, not the basis of the rule.
- Margin: blended gross margin is fine to start; category-level margin if your catalog varies widely.
- Recovery rate: the honest unknown. Start conservative, then replace the assumption with your own measured rate after the first months — it’s on the dashboard.
Why the threshold beats “mail everything” and “mail nothing”
Mailing every abandoned cart torches money on $18 orders; mailing none forfeits the easiest saves in commerce — the $200 cart whose owner just doesn’t read email (that shopper, specifically). The threshold is the whole strategy: free channels chase everything, the paid channel chases only carts where the math is loudly in your favor. Your spend cap then bounds the worst month possible (how caps work).
Judging the results
Two disciplines keep the ROI story true afterward. Count a recovery only when the same shopper completes the same cart within a bounded window — an unrelated order next month is not your postcard’s trophy. And remember some carts return on their own: holdout-style honesty separates what the card caused from what was coming back anyway, which is the number that should decide next quarter’s threshold.
Common questions
›What recovery rate should I assume for a postcard?
Be conservative: low single digits is a sane planning number for the postcard rung, since it deliberately targets carts that already ignored email. At a 3–5% assumption, the threshold math still clears easily on high-value carts — which is the point of the threshold.
›Why compute margin instead of using cart value directly?
Because you spend real dollars against gross profit, not revenue. A $100 cart at 60% margin returns $60 of profit when recovered; the card must be judged against the $60. Skipping this step makes every threshold look twice as good as it is.
›Should the email rungs pass an ROI test too?
No — that is the beauty of the ladder. Emails cost effectively nothing, so every qualifying cart gets them and the ROI question never arises. The threshold exists only for the rung that costs money per piece.