The last unit of a product was sold in the store. At the same minute, an order for that product came from the marketplace. Now you have an order, but no product.
This scenario is the classic example of overselling, the most expensive mistake in multichannel sales. The result is cancellation, return process, customer dissatisfaction, and a drop in the store rating.
In this article, we explained how stock synchronization works, which settings prevent overselling, and how to decide on the synchronization frequency.
Table of Contents
How does synchronization work?
Stock synchronization is the reporting of the current quantity in your system to sales channels. This notification is triggered at specific intervals or when a change occurs.
For synchronization to work, mapping must be established first. There is no equivalent to report for an unmapped product; product matching see the article.
There is a two-way flow. You report stock; the channel sends you orders. When an order arrives, the stock decreases, and the new quantity is reported in the next synchronization.
The time interval in between is the window where the risk arises. Sales made in that window are based on stock not yet reported.
Therefore, synchronization design is built on narrowing that window as much as possible and meeting the remaining risk with a buffer.
Deciding on synchronization frequency
Frequent synchronization increases accuracy but creates load on the system and channel side. Therefore, frequency should be differentiated according to the product group.
Frequency should be high for fast-selling and tight-stock products. This group usually constitutes a small portion of the catalog but carries most of the risk.
Longer intervals are sufficient for slow-moving products with large stock. For a product with hundreds of units, a delay of a few minutes does not cause problems.
Sync frequency should generally be increased during campaign periods. When the sales velocity increases, the risk window must also narrow.
We covered seasonal planning in the seasonal demand planning article.
Buffer stock logic
Buffer stock means reporting a slightly lower quantity to the channel than the actual amount. This difference creates a buffer against synchronization delays.
The buffer quantity can be defined as a fixed unit or a percentage. A fixed unit makes more sense for low-stock products, while a percentage is more logical for high-stock products.
The buffer has a cost: products that could have been sold remain unsold. Therefore, excessive buffering is as harmful as overselling.
The way to find the right amount is through measurement. If the cancellation rate is close to zero, the buffer can be reduced; if there are cancellations, it should be increased.
The buffer decision can also vary by channel; the buffer is kept wider in channels where cancellations are penalized more heavily.
Warehouse segregation approach
A more definitive method than buffering is to define a separate warehouse for e-commerce. The goods in this warehouse are allocated exclusively for online sales.
In this approach, in-store sales do not affect online stock; the risk of conflict is largely eliminated.
The disadvantage is stock efficiency. Allocated goods cannot be sold even if there is demand on the other channel; you need to hold more total stock.
An intermediate solution is to separate the warehouses and set up a fast transfer system between them. When the online warehouse runs low, it is replenished from the main warehouse; you can refer to the transfer slip article.
Which approach is appropriate depends on volume and product structure; we compared them in the warehouse and marketplace stock alignment article.
Synchronization on inventory count days
Stock quantities are temporarily volatile during an inventory count. Synchronization performed during this period may report incorrect quantities to the channel.
There are two solutions. The first is to pause the synchronization of the relevant products during the count.
The second is to perform a partial count and temporarily pull the counted zone from the channel. The partial count system provides this flexibility; warehouse inventory count see the article.
Full synchronization must be performed after processing post-count adjustment slips. If this step is skipped, the inventory discrepancy spreads across channels.
The counting calendar should be planned away from campaign periods; overlapping two intensive workloads multiplies the risk of errors.
Monitoring the variance
It should be measured whether the synchronization is working properly. Three indicators provide a sufficient picture.
The first is the cancellation rate: the percentage of orders canceled due to out-of-stock items. This is the most direct indicator.
The second is the number of synchronization errors. If failed synchronization attempts quietly accumulate, inventory is not updated for days; integration error monitoring see the article.
The third is the quantity difference between the channel and the system. A periodic comparison reveals accumulated variances.
Linking all three indicators to notifications when thresholds are exceeded ensures that the problem is noticed before it escalates.
What to do if overselling occurs?
Despite all precautions, overselling can happen. In this case, acting quickly and transparently limits the damage.
The first option is an alternative source: if the same product is available in another warehouse or store, it can be shipped from there.
The second option is rapid sourcing. If it can be procured from the supplier in a short time, the customer can be notified of the delay and the order can be preserved.
The last resort is cancellation. In this case, informing the customer early causes much less damage than informing them on the delivery day.
Every overselling incident must be recorded and its cause analyzed. Repeating products indicate that the reservation setting needs to be reviewed. Synchronization settings are managed in the e-commerce modulemanaged.
Managing synchronization latency
No synchronization is instantaneous. The latency in between is the main source of the oversell risk.
The first way to manage this risk is to leave a safety margin for fast-moving products.
Not reflecting the last few units to the platform significantly reduces the cancellation rate.
The second way is to differentiate the update frequency based on the product group. Slow-moving products do not have to be updated frequently.
The third way is to temporarily increase the frequency during campaign periods; delays during demand surges are more costly.
When these three measures are applied together, overstock-induced cancellations drop to an acceptable level.
Points to consider
It is essential to have a single source of stock. Stock kept in multiple locations inevitably diverges.
Which warehouse will be reflected on the marketplace must also be clearly defined; summing all warehouses is often a wrong choice.
Opening store inventory to the online channel carries the risk of conflicting with in-store sales.
Return processes also affect synchronization; returned products must not be reflected in stock before they become sellable.
Synchronization errors need to be compiled into a list and checked daily.
An update that fails silently leads to selling with incorrect stock for days.
Frequently asked questions
Is real-time synchronization possible?
It depends on the channel infrastructure. In practice, short-interval synchronization yields results close to real-time.
Is price also synchronized?
Yes, via the channel-based price list; price list management see the article.
Can the reserve quantity vary by product?
Yes, and it should. A single general rate cannot be simultaneously correct for fast and slow products.
Does the B2B channel also feed from the same stock?
It depends on the setup; e-commerce and B2B same stock we discussed the options in the article.
Stock synchronization is the most critical setting of multi-channel sales. When set up well, it remains invisible; when set up poorly, it generates cancellations every week.
We recommend setting a reasonable reserve margin to start with and monitoring the cancellation rate. This single metric will tell you in a short time whether the setting is correct.
Defining a separate synchronization frequency for fast-moving products is also the most effective improvement that can be made from day one.
by consulting with the EQLEM team you can plan your stock synchronization setup.