Once you have a draft, BidBot can tell you how strong it is before anyone else sees it. There are two distinct things you can ask for: an evaluation against the tender's criteria, or a critique of the writing itself.
Both are most useful before formal reviewer input, when there is still time to act on what they surface.
Switch BidBot into evaluator mode
By default BidBot is set up to help you write. The Stress Test skill flips it into evaluating mode, so it assesses your draft rather than extending it.
Open BidBot in the answer editor.
Click the + icon at the bottom of the input bar.
Hover over Skills.
Select Stress Test.
This is the main way to put BidBot into evaluator mode, and worth doing before any of the prompts below.
Turning the skill on does not by itself tell BidBot what to examine. You still need to say what you want stress tested — a score, alignment against the evaluation criteria, structure, constructive criticism, or anything else. The prompts in the rest of this article are written for exactly that.
Evaluate an answer against the tender
BidBot assesses your draft against what the tender actually asks for, and returns structured feedback covering:
Alignment with the evaluation criteria and scoring rubric
Compliance with explicit tender requirements
Clarity, logic and readability
Strength, credibility and relevance of your evidence
Client focus and outcome orientation
Tone, precision and adherence to the word limit
This helps you understand how an answer is likely to score, find weaknesses early, and put your remaining effort where it will change the outcome.
Evaluation prompts that work well
Evaluation prompts are strongest when they are explicit, structured, and written from the evaluator's point of view.
Evaluate this answer against the published evaluation criteria and scoring rubric. Indicate the likely score band (for example minimum, acceptable, good, excellent) and explain clearly what is preventing a top score.
Review this response as if you were a procurement evaluator. Score it against each criterion, justify the score, and highlight where additional evidence or clarity would increase marks.
Assess this answer for compliance and scoring risk. Identify any mandatory requirements that are weak, implied, or missing, and explain the potential scoring impact.
Evaluate how well this response addresses the buyer's stated outcomes and priorities. Where does it feel generic, and how could it be made more client-specific to improve scoring?
Provide a structured scoring breakdown with strengths, weaknesses, and prioritised recommendations to move this answer into the highest scoring bracket.
Ask for constructive criticism
Where an evaluation measures the answer against the tender, a critique interrogates the writing. BidBot focuses on how to improve rather than simply what is missing.
Ask for actionable improvement guidance rather than general feedback - well-structured prompts produce clearer, more usable recommendations.
Provide detailed, constructive criticism on this draft. Identify weaknesses in structure, evidence, and clarity, and propose specific improvements for each.
Critique this answer from a buyer and evaluator perspective. What feels unclear, unsupported, or insufficiently differentiated, and how should it be strengthened?
Review this response as a senior reviewer. Highlight sections that feel generic, over-wordy, or low-impact, and suggest revised approaches.
Identify the top three reasons this answer may fail to achieve a high score, and provide clear recommendations to address each one.
Compare this response against what a strong competitor might submit. Where are we under-selling our capability, and how can this be improved?
Best practice
Turn on the Stress Test skill first, then give BidBot a specific instruction — the skill sets the mode, your prompt sets the target.
Evaluate before sending a draft to a reviewer, so reviewer time is spent on judgement rather than obvious gaps.
Act on the highest-impact findings first — the prompts above are written to rank them for you.
Re-evaluate after significant revisions to confirm the weaknesses have actually closed.
Treat the score band as an indication, not a guarantee. The evaluator is a person.
Related articles
BidBot in the Answer Workspace — refining and regenerating the draft once you know what needs fixing.
Content gaps and flagged issues — gaps BidScript identifies automatically as you draft.
Comments and feedback — bringing human reviewers into the same draft.