Insights > Valuation Operations
The Changing Role of the Junior Valuation Analyst
One of the things I’ve been thinking about as AI capabilities continue to improve is what this means for the next generation of valuation professionals.
There’s been quite a bit of discussion about how AI will make analysts more productive. And I think that’s certainly true.
But I’m not sure productivity is the most important issue.
I think we may be looking at a much more fundamental change in what junior analysts actually do, how they develop their expertise, and ultimately how valuation firms organize their professional teams.
For decades, the traditional career path in valuation has been fairly well established. You start by doing the technical work. You collect information, analyze financial statements, build models, prepare reports, and learn through the review process.
Over time, you develop the experience and judgment required to take on more responsibility.
But what happens when technology can perform a significant portion of the work that has historically been the foundation of that learning process?
I think that’s a question the profession needs to start taking seriously.
The Traditional Analyst Role Was Built Around Execution
If you look at how most valuation engagements are staffed today, there’s a fairly predictable division of responsibilities.
Junior analysts perform much of the initial technical work. They gather and organize financial information, identify normalization adjustments, construct financial models, research market data, and prepare sections of valuation reports.
More experienced professionals review that work, challenge assumptions, evaluate the analysis, and ultimately take responsibility for the conclusions.
There’s a practical reason firms operate this way.
The work needs to get done, and junior professionals generally have the technical skills and capacity to perform it under appropriate supervision.
But there’s also an educational component.
Building a discounted cash flow model teaches you more than how to construct formulas in Excel. It forces you to understand the relationships between revenue, margins, working capital, capital expenditures, risk, and value.
Preparing a financial normalization analysis requires you to examine the underlying business and consider whether historical financial results accurately represent its ongoing economic performance.
And when a reviewer questions an assumption or identifies an error, you learn something that carries forward into the next engagement.
That process is how many of us developed our expertise.
The challenge is that these are also the activities where AI capabilities are advancing rapidly.
From Producing Work to Validating Work
I’ve spent considerable time recently testing newer AI models against actual valuation tasks.
Not just document summarization or basic calculations, but sophisticated financial models with multiple layers of formulas and interdependent calculations.
In one test, I provided instructions for a complex valuation spreadsheet that would ordinarily require substantial time from a capable analyst.
The AI generated a structurally sound model in minutes.
Now, that doesn’t mean the work was finished.
The output still required review. Assumptions needed to be evaluated, calculations verified, and the analytical conclusions considered in the context of the engagement.
But the initial execution was substantially compressed.
And that changes the nature of the work available to a junior analyst.
If a financial model can be generated automatically, the analyst may spend less time constructing it and more time determining whether it’s correct.
If AI can identify potential normalization adjustments, the analyst’s role may shift toward evaluating whether those adjustments are appropriate.
If report sections can be generated from structured inputs, the work may increasingly involve verifying the reasoning, identifying omissions, and ensuring the conclusions are properly supported.
In other words, the role begins shifting from producing work product to validating work product.
That sounds like a logical progression.
But I think it creates a problem we need to examine more carefully.
How Do You Learn to Review Work You’ve Never Had to Produce?
This may be the most difficult question.
We’ve traditionally relied on years of hands-on execution to develop the technical understanding required for professional review.
You learn how a model works by building models.
You learn to identify errors by making mistakes, having those mistakes corrected, and understanding why they matter.
You develop judgment through experience with different businesses, industries, assumptions, and valuation circumstances.
But if AI increasingly performs the initial work, how does a junior professional develop those capabilities?
Consider a new analyst who receives a completed financial model generated by an AI system.
The analyst may be asked to review the formulas, evaluate the assumptions, and determine whether the resulting analysis is reasonable.
But if that analyst has limited experience constructing those models, how confident can we be in the review?
It’s one thing to recognize that an output looks reasonable.
It’s another to understand why the analysis works, where it might fail, and what assumptions could materially change the conclusion.
Those aren’t necessarily skills that develop simply by observing a completed work product.
And I don’t think we can assume that knowing how to use AI tools is the same as understanding the analytical work those tools perform.
In fact, as the systems become more capable and their outputs become more polished, identifying errors or unsupported assumptions may become more difficult.
A model that looks professionally constructed can still be based on flawed reasoning or inappropriate inputs.
The professional reviewing it needs enough underlying expertise to recognize those issues.
That creates an interesting tension.
The technology may reduce the need for analysts to perform certain technical tasks, while the profession still depends on that technical experience to develop competent reviewers.
I don’t think we have a complete answer to that problem yet.
But it’s one that valuation firms, educators, and professional organizations should be discussing now.
The Traditional Staffing Model May Need to Change
There’s also a business implication.
Most professional services firms have historically relied on some variation of a staffing pyramid.
A relatively large group of junior professionals performs much of the technical execution. More experienced professionals provide supervision, review, client management, and final decision-making.
The economics of the firm are built, in part, around that division of labor.
But what happens when the volume of work requiring junior-level execution declines?
A firm may no longer need the same number of analysts to produce a given amount of work.
At the same time, it may need greater capacity for technical review, quality control, and professional oversight.
That could change the balance of experience required across the organization.
I’m not suggesting that junior analyst positions disappear.
There will still be information to evaluate, models to test, assumptions to investigate, and engagement-specific circumstances that require professional attention.
And AI systems will continue to require supervision, particularly when their outputs contribute to professional conclusions.
But I do think the composition of valuation teams could change.
We may see smaller teams performing work that historically required more people. We may also see junior professionals taking on different responsibilities earlier in their careers.
That raises questions about hiring, training costs, career progression, and the economics of professional development.
If fewer junior analysts are needed for execution, firms will need to think carefully about how they maintain a pipeline of experienced valuation professionals.
Because today’s junior analysts are supposed to become tomorrow’s managers, directors, and partners.
And that development doesn’t happen automatically.
Training May Become More Important, Not Less
One possible response is to assume that because AI can perform more technical work, junior professionals simply won’t need the same depth of technical training.
I think that would be a mistake.
If anything, the ability to critically evaluate complex analytical output may become more important as the technology improves.
But the way we develop that ability may need to change.
Historically, much of the training happened through engagement execution. Analysts learned while performing tasks that also contributed to the firm’s billable work.
As those tasks become increasingly automated, firms may need to become more deliberate about professional development.
That could mean requiring junior analysts to construct models themselves, even when AI could generate the same work more quickly.
It could mean using AI-generated analyses as training exercises, asking analysts to identify errors, challenge assumptions, or explain the reasoning behind the results.
And it could mean placing greater emphasis on understanding valuation methodologies, financial relationships, professional standards, and the circumstances that make one analytical approach more appropriate than another.
In some respects, we may need to separate the process of developing professional expertise from the process of producing client deliverables.
That’s a meaningful change.
It introduces a cost that firms may previously have been able to absorb through the normal execution of engagements.
But I think it may be necessary if we want the next generation of valuation professionals to develop the judgment and technical competence required to oversee increasingly sophisticated systems.
The objective shouldn’t simply be to train analysts to use AI effectively.
It should be to develop professionals who understand the work well enough to know when the AI is wrong.
Experience and Judgment Will Still Matter
There’s a broader point here that connects to how we think about professional judgment.
One of the common responses to AI in valuation is that technology can handle the technical work, but human judgment will always remain essential.
I generally agree with that.
But judgment doesn’t develop in isolation.
It comes from experience, technical understanding, and exposure to situations where the answers aren’t always straightforward.
If we change how professionals gain that experience, we also need to consider how it affects the development of judgment.
We shouldn’t assume that because AI can build a model, understanding how to build that model is no longer valuable.
Nor should we assume that every task traditionally assigned to a junior analyst needs to remain part of the production process.
The challenge is separating the activities necessary to produce a valuation from the activities necessary to develop a competent valuation professional.
Historically, those two things have been closely connected.
They may become increasingly separate.
And I think that distinction will have significant implications for how firms manage their people.
Preparing the Next Generation of Valuation Professionals
I don’t think the right response is to resist these capabilities or insist that junior analysts continue performing every task manually.
That wouldn’t make much sense.
The technology is improving, and firms that find effective ways to incorporate it into their operations will have opportunities to improve both productivity and the quality of their work.
But simply introducing AI into existing workflows without reconsidering professional development could create problems over time.
For firm leaders, I think there are several questions worth asking.
What technical skills will junior analysts still need to develop through hands-on experience?
How should we train professionals to evaluate work that may have been generated largely by AI?
Which responsibilities can reasonably shift to less experienced staff, and which require more seasoned professional oversight?
And how do we maintain a strong pipeline of future valuation leaders if the traditional entry-level work changes substantially?
These are not questions with simple answers.
But I think they’re becoming increasingly important.
For decades, the valuation profession has relied on a model where junior professionals develop expertise by performing the work that more experienced professionals eventually learn to supervise.
If AI changes who performs that work, we need to reconsider how the next generation develops the expertise required to supervise it.
The question isn’t simply what junior analysts will do when AI can perform more of the work.
It’s how we develop the next generation of valuation professionals when the work that historically trained them is no longer performed in the same way.
I think that’s a much more important conversation.
Better valuation operations begin with better questions.
Explore how ValueVision is building specialized agents, connected operating infrastructure, and practical advisory services for the future of valuation.