AI Will Replace Analysts. Just Not the Ones You Think.
The analysts who only document meetings and push status updates are in trouble. The ones who build technical skills are about to become untouchable.
A product manager I work with told me something last month that stuck with me. He said, “I used to need an analyst to write my requirements. Now I paste the Slack thread into Claude and get a first draft in 90 seconds.”
He wasn’t being arrogant. He was being honest. And he’s not the only one doing it.
I’ve watched this shift happen in real time across three banking programs over the past year. Business stakeholders are writing their own user stories with AI. Developers are generating their own acceptance criteria. Project managers are building their own status decks. The tasks that used to be an analyst’s bread and butter are being absorbed by the people around them, because AI made those tasks easy enough that they no longer need a specialist.
If that scares you, it should. But only if those tasks are all you do.
Whether you’re a business analyst, systems analyst, data analyst, or any flavor of “analyst” on a project team, the dynamic is the same. I’ve spent the last decade working as a Technical BA in banking and payments, and I’ve never felt more secure in my role. Not because I’m ignoring AI, but because I’ve spent years building the kind of skills that AI actually can’t replace. The skills I break down in The Technical Skills Guide for BAs are exactly the ones that are becoming more valuable right now, not less.
Let me explain why.
The Tasks That Are Already Disappearing
Let’s stop pretending and name them. If your day-to-day consists mostly of these activities, you have a problem.
Writing meeting notes and distributing them. AI transcription tools do this better than you. Otter, Fireflies, Copilot in Teams — they capture everything, summarize it, and distribute it without you lifting a finger. If “documenting meetings” is a line on your resume, remove it.
Drafting basic requirements from conversations. A PM can paste a conversation into any LLM and get a structured set of user stories with acceptance criteria. The output isn’t perfect, but it’s good enough that they don’t need to wait two days for you to write it up.
Creating status reports and dashboards. Jira and Azure DevOps already auto-generate most of what a status report contains. Adding AI on top makes the manual version of this completely redundant.
Translating business language into slightly more structured business language. This is the big one. A lot of analysts describe their job as “bridging the gap between business and IT.” But if your version of bridging that gap is just reformatting what a stakeholder said into a template, AI does that faster and cheaper.
None of these tasks require judgment. They require attention and formatting. That’s exactly what AI is best at.
What AI Cannot Do (and Won’t for a Long Time)
Here’s where it gets interesting. The tasks AI struggles with are the tasks that most analysts avoid because they’re hard.
Navigating ambiguity in a room full of conflicting stakeholders. A VP of Payments says the priority is speed to market. A compliance officer says the priority is regulatory alignment. A developer says both are impossible in the current sprint. No AI is sitting in that room reading body language, understanding organizational politics, and finding a path forward that everyone can live with. That’s a human skill, and it’s rare.
Understanding why a technical decision matters for the business. An AI can explain what an API does. It can’t tell you that switching from batch to real-time processing on this specific payment rail will break the downstream reconciliation process that the operations team in Montreal depends on every morning at 6 AM. That insight comes from domain knowledge and relationships. You can’t prompt your way to it.
If you’re building that kind of domain depth, especially in banking or fintech, Break Into Banking — The Complete BA Guide maps out exactly what you need to know, from SWIFT and ISO 20022 to the regulatory landscape and the interview questions you’ll face.
Challenging a technical design with the right questions. A developer presents an architecture diagram. It looks clean. But you notice the authentication service is a single point of failure, or the data model doesn’t account for multi-currency transactions, or the error handling strategy will silently drop failed messages in the Kafka queue. AI can generate architecture diagrams. It can’t look at one in the context of your specific project constraints and ask the question that saves the team three weeks of rework.
Get Ahmed Squalli’s stories in your inbox
Join Medium for free to get updates from this writer.
Producing deliverables that survive contact with reality. AI can draft a BRD, a data dictionary, or a test plan. It can’t validate any of them against the actual system behavior, the actual stakeholder expectations, and the actual technical constraints of your environment. The analyst who tests their own requirements, who opens Postman and verifies that the API actually returns what the spec says it should, who queries the database to confirm the data mapping is correct — that analyst is irreplaceable. I built API Documentation from Scratch around this exact capability, because it’s the single biggest differentiator I’ve seen between analysts who get replaced and analysts who get promoted.
The Two Types of Analyst in 2026
The split is becoming obvious on every project I join. There are two types of analyst emerging, and the gap between them is widening fast.
The Documentation Analyst. This person’s value proposition is “I write things down accurately and keep them organized.” They’re diligent. They’re reliable. And they’re being automated out of relevance. Not fired, necessarily, but slowly deprioritized. Their role shrinks from “analyst” to “project coordinator” to “we don’t need to backfill that position.” It doesn’t matter if you call yourself a BA, a systems analyst, or a data analyst — if documentation is your entire identity, you’re vulnerable.
The Technical Analyst. This person’s value proposition is “I understand the business problem AND the technical solution well enough to make sure they actually align.” They can read an API contract. They can write a SQL query to validate test data. They can sit in a technical design review and contribute, not just take notes. They use AI as a force multiplier, not a crutch. They’re the ones whose calendar is full because every team wants them in the room.
The documentation analyst is worried about AI. The technical analyst is excited about it. That tells you everything you need to know.
How to Make Yourself Irreplaceable
This isn’t about becoming a developer. I want to be clear about that. You don’t need to write production code or deploy infrastructure. But you do need to build enough technical fluency that you can operate in technical spaces without being a passenger. This applies whether you’re a business analyst, a systems analyst, a data analyst, or a functional consultant.
Here’s what that looks like in practice.
Learn to read and test APIs. You don’t need to build them. You need to understand an API spec well enough to write requirements against it, and test it well enough to verify that what was built matches what was specified. This single skill will change how developers perceive you. You go from “the person who writes tickets” to “the person who catches issues before they reach QA.”
Learn basic SQL. Not enough to design databases. Enough to query them. When someone says “the data migration is complete,” you should be able to run a SELECT query and confirm it yourself instead of taking their word for it. When a defect comes in, you should be able to check the data directly instead of waiting for a developer to look at it tomorrow.
Understand system integration patterns. Know what synchronous vs asynchronous means. Know what happens when a message fails in a queue. Know why idempotency matters for payment processing. These concepts come up in every technical discussion on every modern project, and the analyst who understands them earns a seat at the table.
If you want to build all of these skills in a structured way with real banking examples, The Complete Tech BA Bundle covers the full stack, from API fundamentals to domain knowledge to deliverable templates. It’s everything I wish someone had handed me when I started making the jump from documentation analyst to technical analyst.
Use AI to Go Deeper, Not to Do Less
Here’s the part that most “AI for analysts” content gets wrong. They focus on using AI to save time. Write your requirements faster. Summarize meetings quicker. Automate your status reports.
That’s fine, but it’s a race to the bottom. If AI helps you do shallow work faster, you’re just accelerating your own irrelevance.
The better play is using AI to go deeper. Use it to learn a new domain in days instead of months. Use it to understand a technical concept you’d normally skip over. Use it to generate test scenarios you wouldn’t have thought of. Use it to draft a data mapping and then spend your time validating it against the actual system instead of spending your time creating it from scratch.
The analyst who uses AI to produce more documents is replaceable. The analyst who uses AI to produce better analysis is not.
The Market Is Already Deciding
I’m seeing it in job postings. I’m seeing it in project staffing decisions. I’m seeing it in how teams talk about which analysts they want on their next engagement. The demand is shifting toward analysts who can operate technically, and away from analysts who can only operate procedurally.
This isn’t a prediction. It’s already happening. The only question is whether you’re building the skills that the market is moving toward, or holding onto a version of the role that’s shrinking.
If you’re ready to make that shift, everything I’ve built at The Tech BA Toolkit is designed for exactly this moment. Real skills. Real examples. No fluff.
The analysts who will thrive in 2026 and beyond aren’t the ones who fear AI. They’re the ones who became too valuable for AI to matter.
Ahmed is a Senior Technical Business Analyst with 10+ years in banking and payments. He builds practical guides and tools for analysts at The Tech BA Toolkit.
