Kivora Operations Team · Reviewed by Kivora
Stock and reorder decisions for held and dropship inventory
Updated 2026-08-22 · 5 min read
Quick answer
Set a reorder point from how long a supplier actually takes plus how fast the item actually sells, and hold enough cover for the gap. For dropship ranges you do not hold, the decision is different: your job is to keep the availability you publish honest, which means refreshing supplier stock often enough that a shopper cannot buy something nobody can send.
Stock decisions split cleanly into two problems that get confused with each other. For goods you hold, the question is how much money to tie up and when to reorder. For goods a supplier ships on your behalf, you hold nothing and the question is entirely about accuracy: does the availability on your product page match what the supplier can actually send today.
A store carrying both — which is most stores that started with dropship and added their own lines — needs both answers, and needs to know which range each product belongs to.
Reorder points for stock you hold
A reorder point is lead time multiplied by the rate of sale, plus a buffer for the variability of both. The common mistake is using the supplier's quoted lead time rather than the one you have actually observed, which is usually longer and less consistent.
Measure the rate of sale over a period that reflects how the item really sells. A seasonal product averaged across a year produces a reorder point that is wrong in both directions: too high out of season, too low in it.
- Use observed lead time, not the quoted one
- Measure sales rate over a period that matches the item's seasonality
- Add buffer for variability, not for comfort
- Review the point when either input changes, not on a calendar
Dropship: you are publishing someone else's stock
When a supplier ships on your behalf, the availability on your page is a claim about their warehouse. The risk is not tied-up capital; it is selling something that cannot be sent, which produces a refund, an apology and a customer who does not return.
Refresh supplier stock as often as the supplier will let you, and decide what to publish when a feed is stale or unavailable. Continuing to show yesterday's availability as though it were current is the decision most stores make by default without ever making it.
- Refresh at the highest frequency the supplier supports
- Decide explicitly what to show when the feed is stale
- Prefer showing out of stock over selling what cannot ship
Availability language a shopper can act on
In stock, out of stock, pre-order and back-order mean different things to a customer and to a marketplace feed. Use them consistently and make sure the storefront and any product feed derive from the same underlying number rather than being set independently.
If an item is available but slow, say so on the product page rather than at checkout. A shopper who discovers a two-week wait after entering their card details is the same complaint as a late delivery, arriving earlier.
- One source of truth behind the storefront and any feed
- Distinguish out of stock from pre-order and back-order
- Disclose long lead times on the product page, not at checkout
Cost, price and the margin you actually keep
Keep the cost you paid separate from the price you charge, and make sure the cost recorded against a product is the cost you actually pay today. A stale cost figure quietly distorts every margin report built on it, and the distortion is invisible because the report still produces a plausible number.
Where a platform fee, payment fee or delivery subsidy sits between the two, decide which of them the margin figure is supposed to be net of, and use that consistently. A margin that sometimes includes fees and sometimes does not cannot be compared across ranges.
- Verify recorded cost against what the supplier currently charges
- State whether margin is gross or net of fees, and keep it consistent
- Re-check cost after any supplier price change, not annually
Discontinued lines and the long tail
A catalogue accumulates products nobody buys. They are not free: they dilute search results on the storefront, occupy attention in every stock review, and in a product feed they are items a marketplace processes and sometimes disapproves.
Review the tail deliberately. A product with no sales, no stock and no supplier route is a decision waiting to be made, not a listing.
- Review products with no sales and no stock on a set cadence
- Remove lines whose supplier route has ended
- Keep a deliberately curated range rather than everything a supplier lists
How this guide was prepared
Prepared from catalogue and availability workflows Kivora implements for South African stores selling both held and supplier-fulfilled stock. It is operational guidance and does not constitute financial advice.