BCG Made $3.6B from AI Last Year. Here's What That Means for Your Interview.
BCG now generates 25% of its revenue from AI-related work. That shift is showing up directly in what they test in case interviews. Here's what candidates need to know.
BCG Made $3.6B from AI Last Year. Here's What That Means for Your Interview.
BCG reported that roughly 25% of its $14.4 billion in 2025 revenue came from AI-related consulting engagements. That is $3.6 billion in billings from projects that did not exist as a distinct practice area five years ago.
This is not a trivia fact to drop in a coffee chat. It is a signal about what BCG interviewers are going to put in front of you in a case.
What BCG Partners Are Actually Working On
To understand why BCG's revenue mix matters for your interview, you need to understand what AI consulting actually is at the firm level.
BCG's AI revenue is not primarily from building AI models. The firm is not a technology company. The revenue comes from helping large enterprises make decisions about AI: which use cases to prioritize, how to build the internal capabilities to run AI-enabled operations, how to manage the organizational change that comes with deploying AI at scale, and how to govern AI systems to avoid regulatory and reputational exposure.
The cases BCG partners bring to interviews are drawn from their recent client work. When a partner spent the last six months helping a global insurer prioritize which underwriting processes to automate first, the case they bring to an interview looks like: a client trying to decide where to invest in AI, with limited internal capability and competing internal priorities.
That is a different case than "our client's profitability is declining" — but the underlying analytical tools are the same. Prioritization frameworks, build-versus-buy trade-offs, capability assessment, change management costs. If you can structure those questions clearly, you can handle an AI strategy case.
The Three AI Case Types You Are Likely to See
AI prioritization cases. The client has identified 12 potential AI use cases and needs to decide which three to pursue first. Your job is to build a framework for prioritizing them — typically across two dimensions like value potential and implementation feasibility — and apply it to the data in the case. The analytical structure is identical to a classic portfolio prioritization problem. The vocabulary is new.
AI build-versus-buy cases. The client is deciding whether to develop an AI capability in-house or purchase a vendor solution. This is a variant of the classic make-versus-buy case with an additional dimension: the internal capability to manage and maintain the system over time. Relevant factors include total cost of ownership, vendor lock-in risk, time-to-value, and the client's existing data infrastructure.
AI transformation cases. The client is mid-way through an AI-enabled operating model redesign and the expected value has not materialized. Your job is to diagnose why and recommend a course correction. This is a profitability or performance case at its core — the AI context adds complexity around implementation timelines, change management, and the gap between projected and realized productivity gains.
In all three types, the business frameworks you already know are the right tools. The only question is whether you can apply them confidently without being thrown by unfamiliar vocabulary.
What BCG Is Actually Testing With These Cases
BCG interviews are interviewer-led at most offices and score heavily on communication quality. The AI context in a case tests one specific thing more than others: intellectual flexibility.
Interviewers want to see whether you can engage with a problem domain you may not have worked in before, frame the key questions correctly, and hold a structured argument without relying on memorized content from that domain. The candidate who says "I'm not an AI expert, but the business question here is about prioritization — and the framework for that is..." is demonstrating exactly the skill BCG is testing.
The candidate who says "I don't have a background in AI" and stops there is failing on intellectual flexibility, not on knowledge.
The candidate who tries to demonstrate AI knowledge they don't have — dropping technical vocabulary incorrectly — is failing on credibility.
BCG is a firm that prizes intellectual honesty and precise communication above encyclopedic knowledge. The correct move in an AI case is to locate the business question inside the AI context and answer that question with rigor. That is what partners do on real engagements.
Practice this framework
Work through the Microsoft 2014: Satya Nadella's Turnaround case with AI coaching.
The Vocabulary Floor You Need
You do not need to understand how large language models work. You do need to be able to use the following terms correctly in a business sentence:
Use case means a specific application of AI to a defined business problem. "One use case is automating the claims intake process" is a correct sentence. "The use case for AI is efficiency" is too vague to be useful in a case.
Implementation feasibility covers the data availability, technical infrastructure, and internal capability required to deploy and maintain the AI application. High-value use cases with low feasibility are the most common strategic trap BCG cases surface.
Time-to-value is the lag between investment and measurable return. AI transformation projects typically have longer time-to-value than traditional process improvements because of the training, integration, and organizational adoption requirements.
Governance refers to the policies and oversight structures that determine how AI systems are monitored, audited, and corrected when they produce bad outputs. In regulated industries (financial services, healthcare, insurance), governance is a first-order concern that can block or delay deployment.
Using these four terms correctly, in context, is sufficient vocabulary for most AI-themed cases BCG will put in front of you this cycle.
Why This Matters More This Year Than Last
BCG's AI revenue grew from roughly 10% of total billings in 2023 to 25% in 2025. The firm expects it to reach 40% by 2027. As that practice grows, the ratio of AI-themed cases in the interview pool grows with it. Interviews are drawn from real partner work. Partner work is increasingly AI-related.
This does not mean every BCG case in 2026 is an AI case. Traditional profitability, market entry, and growth strategy cases remain common. It means that the probability of encountering an AI-themed case in your BCG process has increased meaningfully — and that candidates who have thought about how to structure these cases will be better positioned than those who haven't.
The structural adjustment required is small. The analytical tools are the same. The vocabulary floor is achievable in a few hours of reading. The intellectual posture — curiosity over expertise, business framing over technical depth — is exactly what BCG has always rewarded.
Practical Prep in Two Hours
Hour one: Read one case study or article about an enterprise AI deployment in an industry you find interesting — retail, logistics, financial services, healthcare. Read it for the business decisions, not the technical details. Ask yourself: what were the trade-offs? What did success look like? What were the implementation risks?
Hour two: Take a standard profitability or market entry case and practice a version where the client is implementing AI in a core business process. Run through the same analytical structure you would normally use and practice inserting AI-specific vocabulary at the relevant branches.
After two hours, you will have a clearer sense of how AI cases differ from traditional cases — and how little the underlying analytical approach actually changes.
BoardroomIQ's case library includes business scenarios drawn from real corporate strategy decisions, including technology transformation contexts. Start with a case that involves a company navigating a major operational shift and practice framing the AI dimensions when the case invites them.