The Quote Nobody Had Time to Win

The Company

Halden & Vor formulates and applies specialty coatings for marine and offshore equipment — the kind that stops hull steel corroding in the North Sea. A hundred and thirty people, deeply technical. They win work because their engineers actually understand the chemistry, not because they undercut anyone. Sales is treated as an interruption to the real work, and in this company's case that attitude has a real cost.

The Pain

Tenders arrive as dense technical RFQs — substrate specs, environmental conditions, compliance standards, twenty pages of it. To quote accurately, someone has to read the whole thing, match it against the product catalogue and past projects, work out coverage and labour, and put together a compliant proposal. Only two senior engineers can do this well, and they're already committed to delivery.

So tenders just sit. Halden & Vor quotes maybe half of what comes in. The rest go unanswered — not lost on price, just never answered at all. Those are six-figure opportunities, and they're missed because nobody had a spare afternoon, not because the company couldn't do the work.

The Move

This isn't something you wire together on a free afternoon. It crosses too many sources, and the output has to be trusted by engineers who'll stake their reputation on it, so it needs to be properly built.

Here's what that looks like. When a tender lands in the bids inbox or portal, it kicks off a pipeline that pulls from the product catalogue, the spec library, costing rules, and the archive of past won proposals. An LLM reads the RFQ, pulls out every requirement, matches products, drafts the technical narrative, and pre-fills the costing model. What lands in the engineer's queue is a drafted, sourced proposal — every figure traceable back to the rule that produced it.

The senior engineer doesn't read twenty pages cold anymore. They read a draft and apply their judgement where it actually matters. They get through far more tenders without anyone new being hired.

As for who builds this — occasionally, us.

The blind spot

The usual objection is that AI can't handle work this specialised. That's fair, but it's not what it's being asked to do. The engineering judgement stays with the engineer. What the pipeline handles is the reading, the matching, and the assembly — the hours of groundwork that have to happen before the engineer can make a single decision. That part is repetitive and document-heavy, and it doesn't need to be done by the most expensive person in the room.

Their quotes win when they go out. That's never been the problem. The problem is how many proposals two people can physically produce in a week. That's what a build like this addresses.

The pattern

The same approach fits any business where skilled people are spending big chunks of their day on document-heavy assembly before they can do anything that actually needs their expertise:

• A law firm producing first-pass contract reviews against its own clause library and prior matters.
• An insurance broker putting together submission packs from client data, risk schedules, and underwriting guidelines.
• An architecture practice drafting planning-application narratives from drawings, site data, and local-authority policy.