The supplier portal was a category error. Built to give buyers control over their supplier data, it ended up giving the supplier a new form to fill in for every buyer they worked with. The result is fragmentation, not verification. Each buyer's portal contains a partial, often stale, version of the truth. The supplier carries the workload of keeping them synchronised, which they predictably do not. Supplier-led identity is the inversion of the model, and it produces a cleaner outcome for both sides.
Why portals failed
The supplier portal model has been in place for twenty years. It is well understood, widely deployed and structurally flawed.
The model assumes the buyer owns the supplier's data. Each buyer maintains its own portal, with its own supplier records, in its own format. The supplier is asked to fill in the buyer's form, often multiple times across the relationship as renewals, recertifications and updates run. We have argued the broader version of this point in supplier onboarding is broken.
The result is a system where the supplier carries the workload of keeping data current across every buyer they work with. They do not, because the incentive structure does not reward it. A small supplier working with twenty buyers maintains one accurate record, possibly two, and lets the others drift.
The buyer's portal therefore contains data that is partial, out-of-date, and inconsistent with what other buyers hold on the same supplier. The portal solves the buyer's perception of control. It does not solve the underlying data problem.
The portal mindset and what it locked in
The portal model carried three assumptions into AP workflow design.
That the buyer is the centre of the relationship. Every interaction is initiated by the buyer, on the buyer's terms, in the buyer's interface.
That supplier data is per-buyer. The supplier holds different identities, attributes, and records inside different buyers' systems, with no canonical truth.
That verification is a one-off event. Once the supplier is onboarded into the portal, the data is treated as current until the next periodic review.
All three are wrong, and they are wrong in ways that the network era exposes daily.
Supplier-led identity: what changes
The inversion is straightforward. The supplier holds the canonical record. Buyers consume verified attributes from it. Updates flow once, from supplier to network, and are then available to every buyer the supplier works with. The data model is laid out in the supplier identity graph.
The attributes that travel include legal entity name and registration, beneficial ownership, bank details, sanctions status, behavioural baseline, and supplier-defined business attributes such as certifications and insurance cover. Each is verified, not asserted, and carries a freshness indicator. The trust dimension sits in supplier trust scoring.
Buyers no longer rebuild the supplier record. They subscribe to the supplier's verified record and consume changes as they occur. The re-keying tax disappears. The fragmentation problem disappears with it.
What this means for procurement teams
The implication that most concerns procurement teams is the loss of control over the supplier record. The instinct is to read supplier-led identity as a reduction in buyer power.
It is not. The buyer retains full control over which suppliers it engages, on what terms, and against what evidence. What changes is the data architecture, not the commercial relationship. The buyer consumes a verified record rather than maintaining a partially verified one. Control over the relationship strengthens because the data is more reliable, not because more of it sits behind the buyer's firewall.
Procurement teams that adopt supplier-led identity find their workload shifts. The mechanical work of supplier data maintenance falls. The strategic work of category management, supplier strategy and relationship development rises in share. This is the deeper ownership question we explore in procurement versus finance.
Migration patterns
Three approaches dominate, each suited to a different starting position.
Sunset the portal. Run the portal in parallel with the network for a defined window, then retire it. Best for buyers with a small, manageable supplier base and a clean current portal.
Layer the network above the portal. Keep the portal as the buyer-side view but feed it from the network's verified record. The portal becomes a presentation layer, not a data source. Best for buyers with significant portal customisation that is difficult to retire.
Greenfield. New supplier onboarding goes through the network. Existing suppliers migrate on a defined schedule. Best for buyers in a transformation programme who can absorb the operational change.
The future shape of supplier interaction
The portal will not disappear from the language, because the term is broad. What disappears is the buyer-owned, re-keying-tax version of the portal. What replaces it is a supplier-held, network-verified record that buyers consume. The workflow looks similar from the buyer's side. The architecture beneath it is the inversion of two decades of supplier management practice.
.jpg)
