What overselling costs your seller score and Buy Box, using webhooks alongside polling, buffer stock strategy, and why price should differ per channel.
Updating stock by hand while selling the same product on three channels is mathematically unsustainable. This article covers the real cost of overselling, how a synchronisation architecture is built, buffer stock strategy, and the cases that require a different price per channel.
In multi-channel selling the most expensive mistake is not miscalculating margin — it is selling a product you do not have. Overselling does not merely cancel one order; it lowers your seller score, pushes you out of the Buy Box, and starts a chain that takes weeks to recover from.
The real cost of overselling
You have three units left and you sell the product on Trendyol, Hepsiburada and your own site at the same time. If two orders arrive from each channel on the same day, you fulfil three and cancel three. The consequences:
- Failure-to-supply score. On Trendyol this criterion is a direct component of the seller score, and most of its weight looks at the last 30 days — covered in detail in our store management article.
- Loss of the Buy Box. On Hepsiburada a stock interruption drops the listing, and regaining it takes time.
- Customer complaints. A cancelled order usually turns into a negative review.
- Wasted advertising. If you are advertising that product, you paid for clicks on stock you could not sell.
A single stock error therefore costs you in four places. Synchronisation is not an efficiency topic; it is revenue protection.
Why manual management fails
The update load grows multiplicatively with channel count. Selling 500 products across 3 channels means 1,500 operations even at one update a day — and in reality updates have to happen at the moment of the order, not daily.
The practical threshold: two channels and a few dozen products can be managed by hand; a third channel or a few hundred products makes automation mandatory.
Synchronisation architecture
A single source of truth
The first rule is that stock lives in one place — usually an ERP or stock management system — and the channels are fed from it. Setups where channels try to update each other produce conflicts, and it becomes unclear which number is correct.
Webhooks or polling
| Webhook (event-based) | Polling (periodic query) | |
|---|---|---|
| Latency | Seconds | As long as the interval (5–30 min) |
| System load | Low — runs only on change | High — queries even with no change |
| Setup | The channel must support it | Works with any API |
| Risk of missing events | A dropped notification is lost | Caught on the next pass |
The right design uses both: webhooks for instant updates, plus a full reconciliation at regular intervals as a safety net. Systems that rely on webhooks alone keep running on wrong stock, silently, after a single dropped notification.
Buffer stock
However fast the synchronisation, the chance of two orders arriving in the same second never reaches zero. The solution is commercial rather than technical: keep a buffer of one or two units per channel.
Raise the buffer on fast-moving products and take it towards zero on slow movers. A buffer means unsold stock, so it has a cost — but usually a smaller one than a cancellation's damage to your score.
Price synchronisation: the same price is not right on every channel
Stock should be identical across channels; price should not. The reason is simple: the deduction structures differ. Hepsiburada's fixed per-order transaction fee, Amazon's withholding and account fee, and Trendyol's shipping tiers all change the real margin of the same product on each channel. You will find the detailed comparison in our article on commission and profit margin.
The correct design derives a price from the target margin per channel: one cost base is kept for the product, each channel's effective deduction rate is applied, and the price follows. Because this is impractical by hand, pricing rules belong in the synchronisation layer too.
Off-the-shelf integration or custom development
Ready-made integration packages solve standard scenarios quickly: basic stock, price and order transfer. Custom development becomes necessary when:
- You have your own ERP or a bespoke stock system
- Your variant, set or bundle logic is non-standard
- You run different pricing rules and campaign logic per channel
- You need two-way flow with warehouse, production or suppliers
We cover the general decision framework in off-the-shelf or custom software.
Setup checklist
- Decide and write down the single source of truth for stock.
- Clarify the update direction for each channel (one-way or two-way).
- Check webhook support; set the polling interval for channels without it.
- Schedule a regular full reconciliation (at least once a day).
- Define buffer stock rules by product group.
- Derive the per-channel price rule from the target margin.
- Set up alerting for synchronisation failures — silent failure is the dangerous kind.
Common mistakes
- Relying on webhooks alone. One dropped notification means days of wrong stock, unnoticed.
- Keeping no buffer stock. The chance of simultaneous orders is never zero.
- Equalising price across all channels. With different deduction structures, one price means a loss on some channels.
- No alerting. When an integration stops silently, you usually learn about it from a cancelled order.
- Mapping variant products as single items. A mis-mapped size or colour breakdown means selling the wrong variant.
Conclusion
Stock synchronisation looks like an infrastructure topic, but in multi-channel selling it directly determines your seller score, Buy Box position and advertising efficiency. A properly built synchronisation layer turns opening a third channel from a risk into a growth lever.
At Commerslab we build the stock–price–order flow between marketplaces, e-commerce sites and ERP with our own integrations. See our system integrations service or get in touch about your current setup.
