01
Order release on payment
Paid orders drop into Manhattan automatically with the right service level, address validated and gift messages attached.
Outcome: the floor works from one queue, not an inbox.
Manhattan integration specialists
Manhattan API integration and platform bridging
Where Manhattan sits at the centre of a wider stack, the risk is always pick-pack-ship events, stock allocation and cycle counts drifting out of sync. We build the bridge that keeps them aligned.
Fixed-scope engagements, delivered by UK-based integration engineers. Documentation and source code handed over on completion.
- ✓Active WM REST API
- ✓Event streaming
- ✓Enterprise-scale flows
What we can connect Manhattan to
Manhattan holds stock locations, pick tasks, despatches and goods-in. Every other system in the business holds a partial copy of the same data. Our role is to make one the source of truth and keep the rest aligned — through the Manhattan API, webhooks where they exist, and a bridging layer where the other system has no usable interface of its own.
Ecommerce platforms
Release paid orders from your storefront into Manhattan on payment, and write fulfilment status and tracking back so the customer is notified automatically.
Shopify, WooCommerce, Magento, BigCommerce
ERP and order management
Keep Manhattan and the ERP agreed on stock, allocations, purchase orders and goods-in, with one owning system per field.
NetSuite, Dynamics 365, Sage 200, Linnworks
Marketplaces
Route Amazon and eBay orders into Manhattan inside the despatch window, with valid tracking returned to keep account health intact.
Amazon SP-API, eBay
Couriers and shipping
Trigger label and manifest creation from the Manhattan pick-and-pack step and return the carrier, service and tracking to the shipment.
Royal Mail, DPD, Evri, DHL, UPS, FedEx
Accounting and finance
Post stock adjustments, write-offs and goods-in valuations from Manhattan so the ledger and the warehouse agree at month end.
Sage, Xero, QuickBooks
Bespoke, in-house and legacy warehouse systems
A warehouse system built in-house, a 2009 WMS with no support, or a spreadsheet-and-scanner process. Where documentation is thin we reverse-engineer the interface and wrap it in a documented, maintainable layer.
SQL Server, MS Access, SFTP file exchange, hand-held terminal exports
Common Manhattan integration use cases
Each is a defined, testable data flow rather than an open-ended development project.
02
Real-time stock write-back
Picks, putaways, adjustments and cycle counts in Manhattan update the ERP and every sales channel within seconds.
Outcome: oversells and manual stock takes both fall away.
03
Despatch and tracking round trip
Despatch confirmation, carrier, service and tracking number flow from Manhattan back to the order.
Outcome: accurate customer notifications, fewer WISMO enquiries.
04
Goods-in and purchase orders
Purchase orders flow from the ERP into Manhattan; receipts, discrepancies and putaway locations flow back.
Outcome: buying and receiving see the same picture.
05
Returns processing
Return authorisations reach Manhattan; graded outcomes (restock, quarantine, scrap) update stock and trigger the refund.
Outcome: returns cleared same day, stock accurate.
06
Bridging a system with no API
A hosted REST wrapper over an in-house or legacy WMS — read models first, then controlled writes — so it integrates like any modern platform without becoming a second source of truth.
Outcome: a dead-end warehouse system becomes an asset.
How a Manhattan API integration is delivered
Four stages, fixed scope agreed before any code is written. Typical delivery for a single-system integration is two to four weeks.
Stage one
Discovery and data mapping
We document the systems involved, the direction of each flow, and every field that must move — including the awkward ones such as tax, discounts and multi-currency.
Output: written integration specification.
Stage two
Fixed scope and estimate
The specification becomes a fixed-price proposal with a delivery schedule and defined acceptance criteria. No time-and-materials open ends.
Output: costed scope for sign-off.
Stage three
Build and testing
Development against a Manhattan development or sandbox environment and a sandbox of the destination system, with your own data used for acceptance testing.
Output: tested integration and test evidence.
Stage four
Go-live and handover
Controlled cutover with monitoring in place, followed by handover of code, credentials and documentation, and a period of hypercare support.
Output: live integration, owned by you.
Have a Manhattan data flow in mind? We will tell you what it takes before you commit.
Request a scopeManhattan API bridging when there is no direct connector
Off-the-shelf connectors assume both platforms are modern, documented and willing to talk to each other. In practice one side is usually a warehouse system from 2009, a bespoke database or an ERP whose vendor charges per field. Bridging is the work of putting a controlled layer between the two.
That layer normalises data formats, holds a queue so neither platform is overwhelmed, authenticated the way Manhattan expects — OAuth 2.0 — and records every transaction so a disputed record can be traced end to end. It also isolates you from change: when Manhattan deprecates an API version, only the bridge is updated.
Where a platform genuinely has no interface, we work with what exists — scheduled SFTP file exchange, direct database connectors, or a documented REST wrapper we build and host on your behalf.
Included in every bridge we build
- →Error handling with alerting when a transfer fails
- →Automatic retries that respect the platform API rate limits
- →A transaction log for tracing any individual record
- →Idempotency so a repeated event cannot create a duplicate
- →Field-level mapping documentation you can hand to any developer
- →A staging environment that mirrors production
Manhattan integration FAQs
A single-direction integration between Manhattan and one business system is typically delivered in two to four weeks, including discovery, field-level mapping, build against a sandbox, acceptance testing with your own data and supported go-live. Multi-system programmes are phased.
Usually yes. We start by documenting what the system actually does, then expose it through a hosted REST wrapper or scheduled file exchange over SFTP. The wrapper is deliberately narrow — the few operations the integration needs, each validated and logged — and it never holds state that should live in the WMS or the ERP.
One owning system per field, agreed in discovery. Movements propagate on event rather than overnight batch, with per-channel buffers, an origin marker to stop echo loops, and a transaction log so any single unit can be traced from goods-in to despatch.
You do. Source code, credentials, configuration and documentation are handed over on completion, and there is no licence fee for continuing to use it. Monitoring and change support are available under a separate agreement.
Platform APIs are versioned and change on their own schedule. Integrations under a support agreement are reviewed each release cycle and migrated ahead of deprecation dates; for handed-over work the version dependencies are documented so your team can plan.
Other platforms we integrate
Each platform has its own integration guide covering authentication, available endpoints and typical data flows.
