MENU

We built a supplier portal instead of an AI agent. On the left, an agent reads the supplier inbox and guesses each field; on the right, the supplier types it once into a portal on the same record the team, ERP, 3PL and finance read.

We built a supplier portal instead of an AI agent

A brand we spoke to earlier this year was running 300 to 500 samples a month. The brief was PLM.

When we looked at how the team actually worked, every sample request was an email. Every round of fit feedback was an email. Every approval, every chase, every revised date.

There was no system underneath any of it. There was an inbox, and there were people who were good at reading it.

The obvious move in 2026 is to point an agent at that inbox. Pull out the style, the round, the comment and the date, and write it into a record. We could build that in a fortnight, and it would work.

We've faced that choice on more than one account now, and each time we've built a supplier portal instead. It took me a while to be able to say why without sounding like I was talking myself out of revenue.

"You can spend the next five years extracting the data, or six weeks changing where it enters."

Oliver Rhodes, CEO, Nolo

The parser never leaves

An agent reading supplier email is a translation layer, and translation layers are permanent.

Every field it produces is an inference about what somebody meant. Which of the three dates in the paragraph is the ex-factory date. Whether "shipping Friday" means leaving the factory or landing at the 3PL. Whether the quantity in the body overrides the quantity in the attachment.

It gets those right most of the time. The trouble is the other times, and the fact that you can't see which is which without going back to the email.

Then the factory hires someone new who writes differently, and accuracy drops with nothing in the system to tell you it has.

The email is still the source of truth in that setup. Your record is a copy of somebody's prose, and you've taken on a dependency that needs maintaining for as long as you trade with that supplier.

You already have an extraction layer

In January we ran three discovery workshops at a fashion brand we'd been working with for four years.

One was with the person who handles inbound logistics. He processes carrier email notifications by hand, keeps an Excel shipment log past 5,000 lines, and types pre-advice documents into the 3PL portal one at a time.

Someone asked what happens if he's off sick. The room went quiet.

He's the parser. He's been doing the job an agent would do, accurately, for years, and the rest of the business can't see what's on the water without asking him.

So the real question is who does the extracting, and what the business does when they leave.

What changes when the supplier types it

Three things, and the obvious one matters least.

The retyping goes. That's hours back for the ops team, and it's the line that ends up in the business case.

The data arrives structured, because the supplier filled in a field instead of writing a sentence. No one has to guess what was meant.

And the supplier's version and your version are the same row. There was never a second copy, so there's no reconciliation and no afternoon spent deciding which one is right.

That third one is the whole reason to do it.

The portal that writes nowhere

There's a version of this that makes things worse, and plenty of brands have built it.

A login, some forms, and a separate database nothing downstream can read. The supplier dutifully enters data into a system that doesn't feed the product master, so ops export it and retype it anyway.

That's a silo with better manners. It also makes the AI problem harder, because the gap now has a login and looks solved.

A portal only counts if it's a view of the same record everything else reads from. Same product, same PO, same permissions, same audit trail. The supplier sees a narrow slice and can write to a narrower one.

Where does your supplier data come from? Three ways to extract it, each with a problem: an agent parses the email and accuracy drops unseen; a person retypes it into a 5,000-line shipment log and it stops when they're off; a portal that writes nowhere means ops export and retype it. The one way to capture it: a portal on the same record, where the supplier fills the field on your PO and there's one row with no second copy.

What we actually build

At Jaded London, suppliers log in to a portal that sits on the same base the team works in. They acknowledge each purchase order, or reject it with comments. They give their own ex-factory date next to the buyer's, and the portal flags wherever the two disagree.

They book shipments against the PO, carton by carton. Their invoice waits until the Jaded team has accepted the shipment, so finance is matching against a record instead of a PDF and a memory. Samples run through the same login.

At another nutrition brand, new suppliers and co-manufacturers are invited, qualified and approved through the portal. Once they're in, they approve and comment on their own records, and they're prompted before a certification runs out.

Both write to the records the team already works from. Neither creates a second copy.

Suppliers have to want to use it

The build is the easy part. You can't make a factory log in.

At the nutrition brand, adoption was the first risk we named, so the team can work the portal on a supplier's behalf until that supplier takes it on. At Jaded, the decision was to get the process right internally before inviting a single supplier.

Both come from the same place. The portal has to be the easiest way for a supplier to get paid and get answers, or the email comes straight back.

The cost isn't where people expect it

A supplier seat on a system you own costs single digits a month. A supplier licence on an enterprise PLM costs a lot more, which is why so many brands buy the PLM and never roll it out past their own team.

The build is a known quantity too. It's a defined piece of work with a number against it.

Set that against the ops salary currently spent on retyping, and the key-person risk sitting underneath it, and it stops being a technology decision.

Then the AI conversation gets interesting

Once the data is arriving structured and at source, an agent has something worth reading.

Supplier performance across a season. Which factories slip, and by how much. Where the critical path is going to break three weeks before it does. We're running that kind of analysis on client data now, and it works because the data underneath it was captured properly in the first place.

If you're looking at your supply chain and wondering where AI fits, the first question is where your data comes from. If the answer is an inbox, that's the thing to fix. The rest gets much easier afterwards.

What a supplier does in the portal: acknowledge the PO or reject with comments, give their ex-factory date (flagged if it differs), book the shipment carton by carton, and upload the invoice, held until accepted. Each step writes to the same record of product, PO, shipment and invoice, read by the team, finance, the 3PL and the ERP.

What to take from this

01
An agent reading supplier email is a permanent translation layer. Every field it produces is an inference you can't audit without going back to the email.
02
You already have an extraction layer. It's usually one person, and the business finds out how much it depended on them when they leave.
03
A supplier portal only counts if it writes to the same record everything else reads. Anything else is a silo with a login.