Reconciling stock when two channels disagree
Compare every SKU across every channel, see exactly which ones disagree and by how much — before deciding which one is right.
You suspect your channels have drifted apart but you cannot prove it. Checking three thousand SKUs across four channels by hand is not a task, it is a week.
Drift is not dramatic. It is a handful of SKUs, a few units out, accumulated over weeks by something that failed quietly. You will not notice it in a dashboard, and you will not find it by spot-checking, because the products that drifted are not the ones you happen to look at.
You find it by comparing everything at once.
Compare first, decide second
A reconciliation reads what each channel currently believes and lines it up against what Qwikr holds, SKU by SKU. The output is not a fix — it is a list of disagreements with the size of each gap.
Keeping those two steps apart is deliberate, and it is the whole design. A tool that reconciles and corrects in one pass has decided, on your behalf and without evidence, which side was right. That decision is the entire question.
Which direction is a real decision
Here is why it matters, from an account where it happened.
A comparison showed 4,242 of 4,251 SKUs in agreement and nine that were not. The nine were low in Shopify and correct in Qwikr, because they had sold on Amazon and eBay and Shopify had never been told.
The gap had to be closed by pushing Qwikr’s figures out to Shopify. Running the comparison in the other direction — copying Shopify’s figures back in — would have re-added stock that had genuinely been sold, and caused fresh overselling on every channel within hours. Same nine SKUs, same nine gaps, opposite outcome.
That is why the comparison reports and stops. The evidence tells you which way to push; nothing should assume it.
Read the pattern, not the total
The shape of the disagreements usually names the cause faster than the count does:
- A handful of SKUs, each low by exactly what sold — a sync path that is not firing. Something specific and fixable.
- Everything off by a little, in both directions — counting rather than syncing. The warehouse, not the software.
- One channel wrong across the board — that connection, or an allowance that ran out mid-run.
Dry run is the default
Every comparison runs read-only unless you deliberately opt into applying it. Not a preference — a default, because the cost of an accidental run in the wrong direction is measured in oversold orders and damaged seller metrics, and the cost of an accidental dry run is nothing at all.
Stop reconciling stock by hand
One catalogue, six channels, one stock figure that stays right.