Business Analysis

Why business analysis skills matter more — not less — in the AI era

As AI tools automate outputs, the ability to ask the right questions, understand stakeholders, and govern data becomes the differentiator. Here's what that means for BA careers.

Every few months a new wave of commentary suggests that AI will make business analysts redundant. The argument usually runs like this: AI can now generate requirements documents, produce process models, and summarise stakeholder interviews automatically. What's left for the BA to do?

The argument misunderstands what business analysis actually is.

The outputs were never the hard part

A Business Requirements Document is not difficult to produce. Given enough information, a sufficiently capable AI can produce a competent BRD in minutes. The same is true of process models, data dictionaries, and gap analyses.

What AI cannot do — at least not reliably, at least not yet — is determine what information is needed, decide whose version of the truth to believe, identify the unstated assumption that will derail the project in month three, or navigate the political reality that the Head of Operations doesn't actually want this project to succeed.

Business analysis is not document production. It's structured sense-making in environments where the problem is ambiguous, the stakeholders disagree, and the requirements will change before the ink is dry. The documents are evidence that the thinking happened. They are not the thinking itself.

What AI actually changes

AI changes the cost of producing outputs. It does not change the cost of asking the right questions, building trust with difficult stakeholders, or governing the quality of the data that feeds AI systems in the first place.

In fact, as organisations deploy AI tools into business processes, the demand for structured analysis skills increases in several specific ways:

Data governance becomes non-negotiable. AI systems are only as good as the data they're trained on and the processes that govern their outputs. Someone has to map those data flows, identify consent issues, assess quality, and document the lineage. This is BA work.

Automation amplifies bad requirements. Manual processes with unclear requirements produce bad outputs slowly. Automated processes with unclear requirements produce bad outputs at scale. The cost of imprecision goes up, not down, when AI is involved.

Stakeholder management doesn't automate. The BA's most valuable skill — understanding what different stakeholders actually need, as distinct from what they say they want — is a human skill. It depends on reading context, managing relationships, and exercising judgement. AI tools can support this work. They cannot replace it.

The skills that matter now

If anything, the AI era rewards a specific subset of BA skills more than before:

• Data literacy — understanding how data flows, where it's governed, and how it's used in automated decision-making
• Stakeholder elicitation — the ability to draw out real requirements from people who don't always know what they want
• Impact assessment — understanding second and third-order effects of technology changes on people and processes
• Governance and compliance — knowing how regulatory requirements map to technical decisions

These are not new skills. They are the core of the BA role. What's new is the environment they're being applied in — one where the pace of change is faster, the stakes of poor analysis are higher, and the ability to demonstrate these skills clearly is more important than ever.

What this means for BA professionals

If you're building a BA career in 2025, the question is not whether to worry about AI. The question is whether you can demonstrate competence in the skills that AI makes more valuable.

That means producing real deliverables in realistic environments. It means building a portfolio that shows you can navigate ambiguity, manage stakeholder conflict, and govern complex requirements — not just that you understand what those things are.

The analysts who will be most valuable in an AI-accelerated organisation are the ones who can structure the problems that AI tools are then applied to. That role is not going away. It's expanding.

AI Workplace Simulator

Stop describing what you know.
Start showing what you can do.

30 days. 26 verified deliverables across the Junior and Intermediate BA tiers. A portfolio employers can inspect.

More from the blog