From manually processing tenders to submitting twice as many with the same team
A consulting firm with over 200 public tenders won and a capacity problem: each bid consumed between 8 and 14 hours of document work. We developed a custom AI tool that generates technical reports, annexes and full documentation — reducing time per tender by 40%.
−40%
time per tender
−40%
Time per tender
2x
Tenders submitted
+200
Winning projects as base
The challenge
A services firm specialising in public tenders with an above-average success rate in the sector. Their portfolio of winning projects was one of their most valuable assets — but that knowledge was trapped in scattered documents, with no system. Every new tender started from scratch: reading and analysing the specifications (2-4 hours), structuring the technical report (4-8 hours), drafting and completing the required annexes (1-2 hours), reviewing and formatting the final document (2-3 hours). The team was excellent. The problem was that document work left little time for what had the most impact: strategy, differentiation and selecting which tenders were worth pursuing.
The solution
Before writing a single line of code, we spent three weeks auditing the historical tender archive: which projects had been won, why they won, what technical report structures repeated in awarded bids, what language the target public administrations used.
Winning projects knowledge base
We structured and vectorised over 200 winning projects as the tool's knowledge base. This is the core: the AI doesn't generate generic content — it generates content based on what this firm has already proven works for winning tenders.
Technical report generator
The team uploads the specifications. The tool analyses the requirements, identifies the evaluation criteria and generates a structured technical report draft with the required sections and proposed content based on similar previously won projects. The professional reviews, edits and signs — but no longer starts from a blank page.
Automatic annex generator
Standardised annexes (responsible declarations, resource commitment forms, technical sheets) are generated automatically with the file's data. What used to take 90 minutes is now ready in 8.
The results
After the first three months with the tool in production:
- −40% total time per tender (from an average of 11 hours to 6.5 hours)
- The team can submit twice as many tenders in the same period
- The first tender processed entirely with the tool was awarded
- AI-generated drafts require 30% less revision than historical drafts
- The team now dedicates more time to strategic selection and differentiation — the parts that most impact award decisions
The real impact is not just the time saved — it's the compounded capacity. With the same team, the firm can now compete in markets and tenders that were previously out of reach due to resource constraints.
Want results like these?
First consultation is free. We'll analyse your case and tell you what's possible.