Custom integration concept · one page

ZIP Fulfillment: WMS + carrier tracking + Dynamics CRM → a client-facing SLA portal

Prove same-day-by-2pm, 24-hour receiving, 48-hour returns and 2-day reach to every e-commerce brand you serve — live, from the systems you already run.

The problem

ZIP's homepage makes four promises a ShipBob-sized competitor can't easily match: orders in by 2pm ship today, inventory is on the shelf within 24 hours, returns are processed in 48, and two hubs put 76% of the US within two days. Today those numbers live inside the proprietary WMS and the carrier accounts, and a client only sees them when someone pulls a report. If a brand can watch its own SLA hit rate every morning, ZIP's biggest selling point becomes something a client would have to give up to leave.

What changes on day one

  • Each client logs in and sees only their orders — shipped today, same-day SLA %, receipts pending, returns in process, inventory on hand and days of cover.
  • ZIP sees every client on one screen, with the 2pm cutoff clock, dock-to-stock hours and the 48-hour returns queue by hub.
  • An alert fires before a promise breaks — a receipt at 20 hours unstocked, a return at 40 hours, a by-2pm order not yet in a carrier scan.
  • Website quote-form leads land in Dynamics CRM to confirm with an instant acknowledgement and a follow-up task, so no inbound lead waits in an inbox.
  • The homepage can show a live "this week" SLA figure pulled from the same model — proof instead of a claim.

Systems diagram

ZIP'S SYSTEMS INTEGRATION LAYER (AZURE) OUTPUTS Proprietary WMSCarrier trackingStorefront connectorsEDI providerDynamics CRMQuickBooks Online own API · orders, receipts, inventoryUPS · FedEx · USPS (tool to confirm)Shopify / Woo / Amazon (to confirm)SPS / TrueCommerce (to confirm)quotes, clients (to confirm)cost per order (to confirm) Azure FunctionsAzure SQLPower AutomateEntra ID / RLS hourly API pulls, carrier scansSLA clock per order & receiptalerts, form → CRM, tasksclient sees own rows only;ZIP sees all SLA rules: 2pm · 24h · 48h · 2-day Client SLA portalZIP executive viewAutomated alertsWebsite sync Power BI, per-client securedall clients · both hubs · marginTeams / email before a misslive SLA figure on homepage No-licence option: same model published as a secure Azure web app embedded in zipfulfillment.com

As-is vs to-be

AreaAs-is (today)To-be (with the portal)
SLA proofPromises stated on the homepage; verified only when a client asks and someone runs a WMS report.Same-day, 24h and 48h hit rates computed per order and receipt, visible to each client live and to ZIP by hub.
Client visibilityEmail or call to ZIP for order status, inventory and returns; storefront shows tracking only.Client logs in: orders shipped today, inventory on hand and days of cover, returns in process, cost per order.
ExceptionsA missed cutoff or a slow receipt is found after the fact, often by the client.Alert at 20h unstocked / 40h return / by-2pm order with no carrier scan by 3:30pm — before the promise breaks.
Sales & churnDifferentiation rests on the sales conversation; leaving costs the client nothing visible.The portal is part of the service; the live figure on the homepage is the demo.
Inbound leadsQuote form to an inbox; follow-up depends on whoever is watching it.Form → Dynamics CRM record to confirm, instant reply, task assigned, pipeline counted on the executive view.

Phased approach

DiscoveryWeeks 1–3Confirm the WMS API and carrier tools, storefront and EDI connectors, Dynamics and QuickBooks; agree the SLA rule definitions (what counts as "received", cutoff time zones for the second hub); pick Power BI vs web portal.
BuildWeeks 4–10Azure Functions pulls and Azure SQL model; SLA clocks per order, receipt and return; Power BI report with row-level security; alerts in Power Automate; quote-form → CRM flow. Pilot with two clients.
Go-liveWeeks 11–13Roll out to all clients with a one-page how-to; homepage SLA widget with Evolve; 30-day tuning of thresholds and alert noise; handover and admin training.
Also worth knowing: BICS can host a private AI server on your own Azure tenant so client and order data never leaves it (useful for "where's my order" chat on the portal), build Power Apps for dock-side receiving and returns inspection on a tablet, and manage the Azure environment end to end.