Beyond TCREI: power user techniques
Once you're comfortable with TCREI, these techniques will help you get consistently stronger output for complex or high-stakes tenders.
Chain your prompts
For long or complex sections, break the task into a sequence. Rather than asking TenderB to write a 1,000-word response in one prompt, ask it to: First, outline the structure. Then, draft each subsection one at a time. Finally, review the whole for consistency. This gives you more control and produces tighter, more coherent output.
2. Use the critique prompt
After TenderB writes a section, ask it to critique its own output: "Review what you just wrote. What is the weakest part of this response from an evaluator's perspective? How would you improve it?" This self-critique loop often catches issues that a direct rewrite wouldn't.
3. Give the evaluation criteria explicitly
If you have the scoring rubric from the procurement document, paste it into your prompt. Ask TenderB to write directly against the criteria: "The evaluator will score this section on: [criteria]. Write a response that explicitly addresses each of these points."
4. Calibrate against your strongest past work
Keep a small library of your best tender sections in a document. When starting a new section on a similar topic, paste your best example into the E (Examples) field. TenderB will calibrate its output to match your quality bar.
5. Use role-stacking for specialist content
For highly technical sections, use a stacked role prompt: "You are a senior bid writer AND a certified project manager with experience in EU-funded projects. Write from both perspectives." This produces output that balances commercial and technical language.
REMEMBER: Advanced prompting is a skill that improves with practice. The best bid writers in TenderB treat their agents as a living document — updating them after each tender with lessons learned.
Common pitfalls to avoid
Too short a prompt: Vague input produces generic output. Always include T and C at minimum.
Accepting first drafts without review: AI output is a starting point. Your expertise is what wins tenders.
Over-relying on one agent for all tender types: Different procurers have different expectations. Build separate agents for notably different contexts.
Forgetting to update your examples: If your quality standard improves, update the example sections in your agents.