[Illustrative example] Stakeholder review a ringi for a new system with AI objections

This article is not a real customer case. It is an example for illustration only (no approval or funding is guaranteed).

Example setup: an operations manager

Imagine an operations manager at a company planning to bring a system replacement proposal to a ringi. They had a sense of the benefits but couldn't predict what questions the CFO or sales director would raise.

Step 1: Enter the proposal's key points

The manager summarized the time the current process takes, the new system's initial and monthly costs, and the likely decision-makers into a brief for MLP Strategy Council.

Step 2: Draft and review from different perspectives

Three AI models with different perspectives each drafted the proposal — one emphasized cost savings, another focused more on migration risk. Each reviewed the others as Response A/B/C, pointing out missing angles.

Step 3: A chairman AI merges it, then the CFO and sales director object

After the chairman model merged the strongest elements, an AI role playing the CFO raised the concern that “the payback period assumption looks optimistic,” while a role playing the sales director asked about “the impact on frontline sales staff during the transition.” The chairman model drafted responses to both as an objection table.

Step 4: Apply feedback and export for submission

Using that table, the manager added an estimate of dual-running costs during the transition period. The final proposal was exported as a PDF and entered the company's ringi workflow.

Takeaway

What this example illustrates is that anticipating objections from multiple angles before circulating a ringi can improve the quality of your nemawashi. That said, this outcome does not guarantee the ringi will be approved — internal decisions are ultimately made by people, and AI-generated content should always be reviewed by you before submission.