SPS Commerce Launches Visibility Management for POs

What SPS Commerce announced on September 15

SPS Commerce used a September 15, 2026 press release to put three things on the table at once: general availability of its agentic AI layer MAX, a new analytics module called Decision Intelligence, and a new product called Visibility Management. The framing line matters more than the product names. SPS said the announcement combines agentic AI with the intelligence of a network trained on more real trading partner activity than any other, giving brands and retailers the data foundation to move from reactive supply chain management to proactive operations.

The scale claim behind that framing is specific enough to check. According to Investing.com's coverage of the release, the products utilize data from the SPS network, which includes more than 300,000 trading relationships, 750 million annual transactions, and 400-plus system automation partners. That's the dataset SPS is betting its "single view of PO status" claim on. For a reader running EDI for a mid-size manufacturer or retailer, the question isn't whether SPS has the volume. It's whether your own transactions feeding that volume are clean enough to trust the view built on top of them.

What Visibility Management actually connects

Visibility Management pulls ERP, WMS, TMS, and carrier data into a single PO status screen, replacing the current pattern where buyers call logistics to find out where an order sits. Per SPS's own product page, SPS Visibility Management connects to your existing TMS, WMS, carrier data, and EDI without a system overhaul, and captures every milestone from PO creation through DC receipt and store delivery, pulled from your existing TMS, WMS, carrier data, and ASN records, with no ERP integration required.

Investing.com's summary of the launch confirms the same scope from the buyer side: Visibility Management connects ERP, warehouse management, transportation management, and carrier systems into a real-time view of purchase order status. That's a wider net than most EDI-centric visibility tools cast. It means the product isn't just reading your 856 ASNs anymore. It's reading carrier scan events and TMS milestones alongside them, then reconciling the two.

That reconciliation is where the "non-event" concept becomes operationally important. If a PO has shipped but no 214 shipment status update has arrived in the expected window, or a WMS receipt never triggers a corresponding ASN, the system doesn't see "no data." It sees a gap against an expected milestone pattern, and that gap is itself the signal worth escalating. Silence, in other words, becomes an exception.

Why this is an EDI story, not just an AI story

A cross-system "single view" is only as trustworthy as the weakest EDI feed underneath it. If Visibility Management is reconciling TMS load-tender data against your 856 ASN and your trading partner's 214 shipment status, a late or malformed 214 doesn't just delay a status update. It produces a false "non-event" alert, because the system has no way to distinguish "nothing happened yet" from "something happened and nobody told us."

This is exactly the kind of error class SPS is pointing at elsewhere in the same release, though in a different module. A Validation Agent unifies real-time network data, retailer and supplier maps and the history of past trading partner setups with best practices from SPS' most experienced consultants, resolving errors an average of 29% faster; consultants review and approve every recommendation before it's applied. That 29% figure is a useful marker for where AI genuinely helps: pattern-matching against a known-good history of partner setups. It's not a marker for where AI can fix bad source data on its own. An agent trained on your network's transaction history still can't manufacture a 214 that a carrier never sent.

Before you lean on Visibility Management's alerts for anything operationally serious, audit the following against your own trading partner list:

• Which partners' 856 ASNs consistently arrive after the shipment has already left the dock, rather than at time of pack
• Which carriers' 214 shipment status messages have gaps between pickup and in-transit scans longer than your SLA window
• Whether your 940/945 warehouse shipping order and advice pairs reconcile automatically, or whether someone manually chases mismatches today
• Whether your TMS and EDI gateway use the same PO and shipment reference numbers, or whether a cross-reference table sits between them

Any one of those gaps becomes a false exception the moment a vendor's AI layer starts treating missing data as meaningful.

How this compares to TMS-side visibility tools

SPS is building visibility from the retailer-network side; TMS and carrier-connectivity platforms build it from the shipper side, and the two are complementary rather than competing. Visibility Management's edge is that it sits on top of 300,000+ existing trading relationships, so a retailer gets cross-supplier visibility without each supplier building anything new. Platforms like MercuryGate, Descartes, Transporeon, and Uber Freight approach the same "where's my shipment" question from the carrier-contract side, aggregating multi-carrier tracking and tendering data for the shipper directly.

Lighter-weight carrier connectivity tools, including Cargoson, sit in a related but distinct spot: they aggregate carrier EDI and API feeds so a shipper's own TMS layer has clean, timely transport data to begin with. That distinction matters for this story specifically, because Visibility Management's PO view is only as current as the 214 and 940/945 data flowing up from wherever your carrier connectivity actually lives. SPS can build the most capable reconciliation layer in the market, and it still inherits whatever timing gaps exist in your carrier integration underneath it.

What EDI and IT teams should do now

The takeaway for trading partner compliance teams

The real work here isn't adopting SPS's new modules. It's making sure the EDI transactions feeding them, your 856s, 214s, and 940/945 pairs, are complete and arrive on schedule, because that's what any vendor's AI visibility layer is actually reading underneath the dashboard. SPS isn't alone in pushing this. Competing EDI providers are building similar exception-detection tooling into the same window, which means data quality audits that used to be a once-a-year compliance exercise are becoming the prerequisite for every new AI feature your EDI vendor ships. Treat the next few quarters as a forcing function: get your 214 timing and 856 completeness in order now, before you're troubleshooting why a vendor's AI flagged a shipment that actually arrived on time.