We believe public procurement should be decided by what a supplier can deliver, not by how many hours it can spare for paperwork. The rules exist for good reasons. The work they create has become a barrier of its own: capable suppliers stay out of competitions they would have won, and buyers see fewer bids than their market could give them.
Winly is an applied AI company that wins public tenders together with partners, with AI systems purpose-built for procurement and developed alongside bid experts.
Our partners are qualified suppliers who could bid on their own, and often have. What changes with Winly is where their hours go.
Winly
Winly's bid experts and AI systems carry the weight of the work. Winly finds the competition and reads every annex. Winly builds the requirement matrix, prices the offer, writes the response criterion by criterion and assembles the submission. Between competitions, the Winly app keeps each partner's market in view: contracts about to expire, notices matched against what the company delivers, and a daily brief on both.
The partner
The partner brings what only the partner can: confirmed capacity, references, valid certificates and the plain facts of what the company delivers. The partner decides whether to bid, what to offer and at what price, and whether the bid goes out. Nothing is submitted without a named person at the partner approving it.
The partner's own people stay on the work the company is already paid for, and the response they sign is written against the criteria the buyer scores. What the partner knows about its trade, its buyers and its competitors is built into the systems along the way. The next bid starts from more than the last one did.
An answer from the buyer can reopen a requirement Winly closed weeks ago, and the price standing on that requirement stops being true. Winly reworks what that answer touched, not the whole bid.
Before
Before a notice exists, Winly tracks contract expiry across the public market and names the competitions likely to reopen. Winly matches what your company actually delivers against the predicted procurements.
Delivers
During
Winly owns the terms and the formal grounds that reject a bid before anyone reads it. Across every annex, Winly builds the requirement matrix, classifies the deviations, and proposes what to redact.
Delivers
Long before submission day, Winly owns capacity, delivery model, subcontractors, and the certificates that must still be valid. From the people and subcontractors your company has already confirmed, Winly builds the capacity claims.
Delivers
Winly completes the price form in the buyer's own format and returns it for your approval. Winly prices no line without confirmation from operations and builds the cost stack underneath the form.
Delivers
While the competition runs, Winly drafts the questions to the contracting authority and holds them for a named approver. Winly reads every clarification the buyer publishes and reopens the affected sections.
Delivers
Winly writes the response one award criterion at a time, not as a general pitch. For the solution description and the reference cases, Winly draws on projects your company has already delivered.
Delivers
Requirement by requirement, Winly checks what your company can actually deliver and marks the gaps in plain language. Winly leaves out any claim no one inside the company has confirmed.
Delivers
After
Last in the process, Winly assembles the bid in the formats the buyer requires and checks every document against the requirement matrix. Winly hands over the finished package only after a named person at your company has approved it.
Delivers
Every vendor draws on the same language models through the same interface, which makes the model the least distinguishing part of any bid a buyer opens. What separates Winly is what the systems learn from.
Bid experts lead the work and correct the systems where they fall short, competition by competition. Each correction stays in the system, and the same mistake does not come back in the next bid.
A submitted bid comes back with a verdict: the buyer's score on each criterion, and the reasons behind it. That verdict is the one measure of what wins a public tender, and it is written only for bids that were delivered.
Behind both sits a structured record of the public market: notices, awards, expiring contracts and case law, joined to what each partner has bid and won. Every competition adds to it.
This is what Winly is building, one competition at a time: AI systems trained on one task, by the verdicts of the buyers who decide it.
We take on a limited number of partners who supply the Norwegian public market. Every partnership starts on a real competition with a real deadline, not on a demo.