Picking Your First AI Use Case: Why Alignment Beats Ranking
There is no objectively right first AI use case. There is only the one your firm is aligned to succeed with.
By the time a firm gets to the question of 'which AI use case should we start with,' they have usually already done something important: they have decided to move. That decision is harder than it looks. The real challenge that follows is not selecting from a menu of options — it is choosing the one that gives your firm the best chance of building organizational confidence in AI.
This article will not give you a definitive list of the best first AI use cases for accounting firms. The research — and the firms that have gotten this right — will tell you that context matters more than category. What we will give you is a framework for making that choice well, grounded in where your people, leadership, and systems are today.
What the Research Actually Says About AI Adoption
Most firms approach AI adoption backwards. They start with the tool and work backwards to the problem. The data from firms that have scaled AI successfully tells a different story.
Just 5% of companies achieve substantial value from AI, while 60% report no material value at all. Source: BCG, 'Why AI Change Is Actually a People Change,' 2026
The difference between firms that extract real value and those that do not is rarely the technology. BCG's 2026 research is direct about this: the majority of AI value is derived from human and organizational factors, not the software itself.
70% of the value from AI transformation comes from people-related action, not technology-related action. Source: BCG, 'Why AI Change Is Actually a People Change,' 2026
This reframes the use case selection question entirely. The best first use case is not the one with the most impressive AI capability. It is the one where your people are most prepared to adopt it, your leadership is most committed to seeing it through, and your data and systems can support it without a major overhaul.
Firms that try to run too many AI initiatives at once consistently underperform those that pick a narrow focus and execute it well.
Successful companies prioritize just three or four AI use cases on average, compared with six or seven for less-successful organizations. Source: BCG, 'Why AI Change Is Actually a People Change,' 2026
The implication for accounting firms, especially mid-market and regional firms, is significant. The pressure to 'keep up' with AI often pushes firms toward a broad, unfocused adoption approach. The evidence suggests the opposite is true: fewer use cases, executed deeply, produce better outcomes.
The Alignment Framework: Three Pillars Before You Pick
Before selecting a use case, assess where your firm stands across three dimensions. These are not gates to pass — they are inputs to your decision. Every firm will have gaps. The goal is to select a use case that maps well to your current strengths, not one that assumes capabilities you are still building.
Pillar 1: Departmental Support
Which part of your firm is most ready to work with AI today? Readiness is not enthusiasm — it is the combination of data access, workflow clarity, and staff capacity to test and iterate. A department that runs on well-documented processes and clean data is a better starting point than one that moves faster but operates on tribal knowledge and fragmented systems.
Questions to ask:
Where do your staff spend the most time on repetitive, rule-based tasks?
Which team has the most standardized workflow documentation?
Which manager is most likely to champion the work, not just approve it?
Where would a visible win generate the most organizational confidence?
Pillar 2: Leadership Commitment
AI initiatives without active leadership commitment do not fail loudly. They fade. A pilot that shows promise but lacks a sponsor who can protect the time, the budget, and the team from competing priorities will stall between phases. This is not a technology problem — it is a governance problem.
The right first use case is one where a leader in your firm has both the authority and the genuine motivation to see it through. That means they are willing to have the conversation when results are not immediate, willing to push back on the team when adoption stalls, and willing to measure success honestly rather than declare victory too early.
76% of executives believed that employees were excited about AI, yet only 31% of employees actually felt that way. Source: BCG, 'Why AI Change Is Actually a People Change,' 2026
This gap matters when choosing a use case. Leadership confidence in AI adoption and employee readiness for it are not the same thing. A use case selected because leadership is excited about it — without accounting for where staff actually are — sets up a difficult rollout.
Pillar 3: Technical Readiness
Your first AI use case does not need a perfect technology environment. It does need a workable one. Before selecting a use case, understand what data will feed it, where that data lives today, and what it would take to make it accessible to an AI system. Firms that skip this step often discover mid-pilot that the data they need is either unavailable, unreliable, or locked in a system that cannot be connected without significant IT work.
Minimum technical questions to answer before committing to a use case:
Is the relevant data digital and centralized, or distributed across systems?
Is there a clean, consistent data format for the AI to work with?
What does the integration path look like with your existing practice management or ERP software?
Is there internal IT capacity to support the pilot, or will it require external resources?
AI Implementation Is a Skill, Not a Switch
One of the most persistent misconceptions in AI adoption is that deploying a tool is the hard part. It is not. The hard part is building your firm's capacity to use AI well — and that capacity is built through deliberate practice over time, not a one-time implementation.
AI implementation is a skill. Like any skill, it requires practice, iteration, and the willingness to fail small in order to learn.
Research on technology adoption supports this directly. A study published through the University of Nottingham found that AI literacy — the ability of staff to understand, evaluate, and work effectively with AI tools — is a learned competency, not a fixed trait. Organizations that treated AI capability as something to be developed through structured learning and iteration outperformed those that treated it as a binary state: either you have an AI tool or you do not.
This has practical implications for your first use case:
Expect the first 90 days to be about learning, not delivering. Productivity gains come after the team has internalized the tool into their workflow.
Build feedback loops from day one. Staff should have a clear way to flag when the AI is wrong, when it is unhelpful, and when the workflow design needs adjustment.
Treat mistakes as data. An AI that produces an incorrect output is not a failed tool — it is a training signal that makes the next iteration better.
Train before you expect performance. Firms that deploy AI without structured training consistently see lower adoption rates and higher staff frustration.
Only 36% of employees say they have been trained on the skills needed for AI transformation. Source: BCG, 'Why AI Change Is Actually a People Change,' 2026
The gap between firms deploying AI tools and firms investing in AI skill development is wide. The firms closing that gap are not the ones with the most advanced technology stack — they are the ones that treat every employee who works with AI as someone who needs active development, not just access.
How to Evaluate a Use Case Candidate
Once you have assessed your three pillars, you can evaluate specific use case candidates against a consistent set of criteria. The goal is not to find the 'best' use case in the abstract — it is to find the one that best fits your current organizational state.
For each candidate use case, ask:
Is this use case solving a problem that staff actually feel? Adoption is faster when AI addresses a pain point the team already experiences as painful.
Is the success criteria definable in advance? 'Better' is not measurable. Define what good looks like before you start.
Is the feedback loop short? Use cases where you can see results within days or weeks beat those where outcomes take months to measure.
Is the risk bounded? Your first use case should not touch client-facing deliverables without a human review layer. Start internal, then move outward.
Is there a clear owner? Someone on your team should be accountable for the outcome — not just the IT lead who manages the tool, but a business owner who owns the result.
A Simple 90-Day Implementation Framework
Once you have selected your use case, the structure of the first 90 days matters as much as the selection itself. Most firms that fail at the pilot stage do so not because the use case was wrong — but because there was no plan for the organizational work that makes the technology stick.
Days 1–30: Foundation
Define the success criteria and baseline metrics before turning anything on.
Identify and brief the 3–5 staff members who will run the pilot.
Audit the data that will feed the AI system — clean it, standardize it, document it.
Set up a weekly check-in cadence with the business owner and pilot team.
Days 31–60: Controlled Rollout
Deploy to pilot team only — not the full department.
Run structured feedback sessions every two weeks.
Document every workflow change the AI requires — this is your implementation guide for broader rollout.
Track both productivity metrics and staff experience metrics (friction, confidence, error rates).
Days 61–90: Assess and Decide
Measure outcomes against the baseline you set on day one.
Review the feedback documentation and identify the top 3 workflow improvements needed.
Make a formal go/no-go decision: expand, iterate, or change direction.
Brief leadership on findings and present the case for next steps.
The goal of your first 90 days is not to deliver ROI. It is to build organizational confidence, identify real-world friction, and develop the internal knowledge to execute the next initiative better.
Where This Fits in Your AI Strategy
Use case selection is not a one-time decision. It is the beginning of a capability-building cycle. Each use case you run — whether it fully succeeds or teaches you something unexpected — adds to your firm's institutional knowledge about how to implement AI effectively.
That is why the roadmap matters as much as the first pick. Firms that approach AI as a series of isolated projects rarely develop the organizational muscle to scale. Firms that treat each use case as a step in a longer journey — building toward a connected, strategy-aligned AI program — are the ones building durable advantage.
In the next article in this series, we will address what comes after early adoption: building the governance and oversight structures that allow AI to operate at scale without introducing unacceptable risk. That includes policies, review frameworks, and the organizational accountability structures that mature AI programs require.
Next in the Series: Article 6 — Governing AI at Scale: Policies, Oversight, and Accountability for Accounting Firms
Publishing November 11, 2026
References
1. BCG. (2026). Why AI Change Is Actually a People Change. Boston Consulting Group.
2. Whittlestone, J., Nyrup, R., Alexandrova, A., & Cave, S. (2019). The Role and Limits of Principles in AI Ethics: Towards a Focus on Tensions.
ACM Conference on Fairness, Accountability, and Transparency. University of Cambridge / University of Nottingham.
3. Longoni, C., Bonezzi, A., & Morewedge, C. K. (2019). Resistance to Medical Artificial Intelligence. Journal of Consumer Research.
Oxford University Press. https://doi.org/10.1093/jcr/ucz013
4. Duan, Y., Edwards, J. S., & Dwivedi, Y. K. (2019). Artificial intelligence for decision making in the era of Big Data — evolution, challenges and research agenda. International Journal of Information Management.
About This Series
"Building Your AI Strategy: A Practical Guide for Accounting Professionals" is an 8-part series published monthly by Rick White at Amarissen. The series is designed for managing partners, operations leaders, and IT directors at accounting firms navigating AI adoption. Each article builds on the last, from strategy and readiness through governance, culture, and measurement.
Comments