All work

Multi-channel sellers — under NDA

Getting a seller's own site and the channels they sell through to agree on stock.

Sector
E-commerce and marketplaces
Engagement
Duration — to confirm
Scope
Integration across marketplace, inventory and fulfilment
Stack
Node.js middleware · marketplace and carrier APIs · AWS

01The problem

Stock was true in four places and correct in none of them.

Selling in more than one place doubles the work unless the channels share a source of truth. Without one, the business is choosing every day between overselling and holding buffer stock it does not need.

Narrative drafted from the brief — replace with the engagement detail

02What discovery found

  1. Oversells clustered around a small number of fast-moving lines, not the whole catalogue.

  2. Each system had a different idea of when an order became committed.

  3. The reconciliation spreadsheet was load-bearing, and nobody had written down its rules.

03What we built

One stock figure, published outward, with fulfilment reading back into it.

  • Marketplace connectors

    Mirakl and the channels alongside it, fed from one source rather than each maintained by hand.

  • Inventory

    Linnworks as the system of record, with the commitment rules made explicit.

  • Fulfilment

    ShipStation closing the loop, so despatch updates stock rather than a person doing it later.

04 — What changed

Qualitative — add figures when you have them

  • One stock figure

    Every channel reads the same number, so the buffer stock that covered the gap is not needed.

  • The spreadsheet is gone

    Its rules are in the system, which means they can be read, questioned and changed.

  • Adding a channel is routine

    The connector boundary makes the next marketplace a configuration job rather than a project.

Got something shaped like this? Let's talk.

Discovery is free — you get a written proposal either way