Help Center

Technical evaluators

Last updated: 10/10/2026

Technical evaluators are the people (internal or external) who answer bidders' questions and score the technical part of the proposals. They are managed in Technical evaluators and work from their own portal, without needing a Sourced account. The full evaluation flow is in Proposal evaluation.

Manage the pool

  • Create: name, email and title/specialty.

  • Edit, disable and restore: disabled evaluators can't be assigned to new tenders (you can include them in the list with the toggle).

  • From each evaluator you can see the tenders they are assigned to.

  • When inviting evaluators to a tender you can choose them from the existing pool or add new ones on the spot; in that case they are added to the pool for future tenders.

What the evaluator sees and doesn't see

The evaluator only sees what they need to weigh in on the technical side, and nothing about the commercial side:

  • Sees: the bid specification and the tender documents, the technical notes of each proposal, the criteria you share and the proposal attachments you release one by one.

  • Doesn't see: prices, payment terms, commercial notes, the commercial analysis or any attachment you haven't released. Commercial data stays hidden so as not to bias their evaluation.

Releasing proposal attachments

Files uploaded by the supplier start hidden from the evaluator: until you release them, they are neither listed nor downloadable. To release them, open the proposal comparison, Documents section, and click the eye icon on each file you want them to see. You can hide it again at any time, even while the tender is running.

Check each file before releasing it. The type the supplier declares on upload (technical, commercial or other) is only a hint and is not always right: priced spreadsheets often arrive tagged as “other”. What decides what the evaluator sees is the eye icon, not that type.

You control which criteria are shared with the evaluators (usually only the technical ones; price and terms are left out) and you can assign each evaluator only some criteria, since they don't need to weigh in on everything.

How the evaluator works

When you invite them to a tender they receive an email with their personal link, which tells them what they will be able to do depending on the stage: if the tender is already under evaluation, score the proposals on the criteria you shared with them and see the specifications and the proposal documents you released; if not yet, answer the technical questions you route to them, see the specifications and, when offer reception closes, score. The first time they get in, they verify their identity with a 6-digit one-time code sent by email in your organization's language (without creating an account); after that they stay signed in for a year. They enter their inbox, where they see:

  • Pending technical questions: the evaluator sees each question —along with any files the bidder attached (drawings, technical PDFs), which they can download— and a draft answer suggested by AI, which includes a confidence level. They can use that draft as a base, edit it or discard it, and attach their own files to the answer. They save drafts and, when they send the answer, it stays pending review by the buyer, who can publish it as is —with its attachments— with Approve & publish, edit it before publishing, or return it with feedback. The published answer is credited to the evaluator: it shows as "Answered by (evaluator) · approved by (whoever approved it)". If another evaluator has already taken the same question, the system warns them.

  • Proposals to evaluate: when the evaluation opens, the evaluator scores each proposal on the criteria assigned to them, with a score from 0 to 100, a confidence level, a mandatory justification and detected risks. The AI offers them a technical summary and a suggested score (limited to their criteria) that they can adopt or ignore.

Progress in the inbox

In the inbox, each tender shows how many evaluations the evaluator has done and how many are left. It only counts the proposals in play in the current round: those of that round and the offers from a previous round that you kept for a supplier who didn't quote again. Versions replaced by a newer offer don't count. A score flagged for review counts as pending until they confirm it.

When you open a new round and the supplier only changed prices or terms, the evaluator sees their previous score as saved, with a note in gray ("Score from round N"), and has nothing to do. If something technical changed, the score shows in amber so they confirm it by saving again or correct it. Once they save it, it becomes their score on the new offer. What counts as a technical change is explained in Proposal evaluation.

What's good to know

  • The justifications and scores stay in the tender's record and are consolidated in the scorecard together with the AI evaluation. When an evaluator submits their scores, those scores replace the AI's on the criteria they evaluated and the weighted total is recalculated.

  • If you need to remove an evaluator from a tender, you can revoke their access from that tender's detail.

  • The invitation link doesn't expire: it only stops working if you revoke their access. What expires is the session, after a year; to get back in they open the same link and request a new code. Each code is valid for 10 minutes and allows 3 attempts; if it didn't arrive, they can resend it from the same page.