This is an illustrative training example, not a named client engagement.
Situation
A 150-person advisory firm believes proposal production is too slow. Leadership asks whether generative AI can write proposals automatically.Diagnostic scope
Start: qualified opportunity requiring a formal proposal. End: approved proposal ready for client submission.Stakeholders
Sales lead, bid/proposal manager, delivery subject-matter expert, commercial reviewer, senior partner and operations/knowledge owner.Evidence
AI-led interviews show that different teams describe the process differently. Consent-based screen observation with a bid manager reveals significant time searching shared drives, copying information between documents and revisiting drafts after missing inputs are discovered.Findings
- The largest delay is collecting missing inputs and waiting for review, not first-draft writing.
- Reusable case studies and biographies are poorly organised.
- Commercial review detects missing scope assumptions late.
- Subject-matter experts answer similar questions repeatedly.
- Drafting is a credible AI-assistance opportunity once inputs are standardised.
Opportunity backlog
- proposal intake checklist;
- approved reusable content library;
- AI-assisted first draft with consultant review;
- earlier, targeted review routing.
Prioritisation
The consultant recommends intake standardisation and an approved content library as prerequisites, then an AI-assisted drafting pilot.Business case
The consultant models proposal volume, drafting time, search time, rework and approval delay outside Iris. Win-rate uplift is treated separately rather than assumed.Report Composer
The current-state map, evidence-backed findings, opportunity sequence and consultant-created business case are assembled into the final report using Iris Report Composer.Executive recommendation
Fix the information and review workflow first, then introduce AI-assisted drafting against approved content. That sequence addresses the dominant delays while keeping commercial judgment with the consulting team.