3PL WMS: every client's stock, witnessed separately

A 3PL’s real product is trust

Your client cannot see their pallets. They see a number on a screen and an invoice, and they believe both because they have to. The moment a count comes back different from what the portal said, the relationship is on a clock.

DockingWare is built so that the number the client sees is the number the floor wrote. One ledger per company, one row per movement, a reference and a user on every row. No parallel spreadsheet, no nightly job that corrects the balance. When a client asks why their stock changed, the answer is a list of rows with names on them.

Separation you can test, not just promise

Multi-tenancy in DockingWare is not a filter on a report. The company is part of every operational table, every query is scoped by it, and the backend test suite sweeps the API endpoint by endpoint to assert that a request for another company’s record comes back not found. Cross-tenant leakage is a red build, not an incident report.

Access follows the same line. Roles and permissions are per company. Audit logs record who did what. Sign-in works with a password or with Google, and Google can only sign in an account that already exists. Nobody self-provisions their way into a warehouse.

The floor, on handhelds and at benches

Inbound is a guided receive against the purchase order: scan the product, confirm cases and pieces, capture lot and expiry when the product is lot-tracked, then palletize and put away. Received on the dock and available for sale are different states, and stock does not become sellable until putaway says so. Expired lots are refused. Short shelf life demands an acknowledgment.

Outbound runs in the modes a mixed 3PL floor actually needs: walk the pick on a handheld, work from a keyboard-first station console at a bench, or pick to tote and hand off. Business-to-business packs verify every line by exact scan before they can close. No scanner on the desk means the handheld, with no bypass. Each confirmed line records who verified it and when.

Locations are labeled in Code 128 from Location Master and scan back through the same verify contract the handheld uses. Pallets carry license plates. Packaging materials are stock, consumed on pack. Parcel labels are bought at the bench (rates, buy, void, buy again), with carrier cost and billed amount stored apart, per tenant.

What clients bring, the warehouse absorbs

Clients arrive with their own barcodes, their own case packs, their own marketplace stickers. A product can hold alternate barcodes that are unique across the whole catalog, so a relabeled unit scans home to the real SKU on every surface, inbound included. Case conversions are per product, and a receipt of fifty-five pieces at fifty-four per case reads as one case and one piece, never as one case.

Client ecommerce channels connect the same way a brand’s would: orders in, availability out, fulfillment back, with a reconciliation that names drift instead of hiding it. Marketplace removals and customer returns without a local order get their own document and post as channel inbound, so a client’s return-rate report stays about returns.

Who this is for

Small and mid-sized 3PLs, from a handful of clients to a few dozen, running out of spreadsheet or running a legacy system whose numbers nobody fully trusts. Operators who want the floor on handhelds today and the client looking at the same ledger the floor wrote.

It is not, yet, for a 3PL whose first requirement is automated client billing. That is coming. It is not here, and the product will not tell you otherwise.

What you can verify

Each capability on this page is tied to a tagged release with a rollback plan. The changelog lists them by version, in the words they shipped with.