10 Steps to Process Improvement with Generative AI

Key Takeaways from This Webinar

In this session, Mark Beller, Knowledge Engineer at SoftEd, explores how generative AI can transform business process improvement (BPI) using a practical 10-step framework. Hosted by Tiffani Aino, the webinar demonstrates how analysts can accelerate process analysis, improve accuracy, and collaborate with AI responsibly.

1. Generative AI amplifies—not replaces—human analysis. It accelerates documentation, modeling, and idea generation while leaving judgment and validation to humans.
2. The “10-step roadmap” still applies. From identifying processes to implementing change, each phase can be enhanced with AI-assisted brainstorming, validation, and automation.
3. AI speeds up the boring work. Tasks like drafting process lists, comparing steps, or merging data sets become minutes instead of hours.
4. Private AI = safer AI. Use on-prem or local large language models (LLMs) to protect sensitive business data while leveraging the same capabilities as cloud agents.
5. Use AI as your process co-analyst. Treat chatbots as collaborators—ask, challenge, and refine responses to reach better insights.
6. Hallucinations can be helpful in early ideation. Use unexpected outputs as creative prompts for brainstorming—but always verify before adoption.
7. Agility and iteration are key. AI enables faster feedback loops, supporting more agile and incremental improvement cycles.
8. Visualization is evolving. AI can generate use cases, diagrams, and models from plain text using tools like PlantUML—perfect for rapid iteration.
9. Context matters more than syntax. Focus on prompting for clarity and outcomes instead of perfect wording—dialogue drives quality.
10. Continuous improvement never stops. Embedding AI into everyday BPI fosters constant feedback and efficiency across teams.