In Plato's "Allegory of the Cave," prisoners mistake shadows for reality — until one escapes and realizes the truth lies beyond the cave wall. At first, this awareness is destabilizing. But it is also liberating.
Since ChatGPT burst into the mainstream a few years ago, many responded to AI advancements in accounting like that cave: partners and firm leaders squinted at generative AI and dismissed it as hype. The skepticism was strong. "Is this thing going to stick around? Should I even care? Do I really need to do anything?" was a common refrain.
But the good news? The tide is changing, and the predominant sentiment is shifting as a profession, from one of skepticism to acknowledgement. Today, I see increasingly more leaders acknowledging that AI will significantly shape the future of the profession.
This is a moment to celebrate because the hardest place to create change is when people don't believe change is needed. If they're content with shadows, there's no conversation to be had.
And yet, that awakening has brought a new problem: paralysis.
"I want to do something, but I don't know where to start."
"There are so many vendors that have AI, I don't want to choose wrong."
"I can't justify the cost until I know the ROI."
This kind of stalling is everywhere. Here's the silver lining: paralysis is an easier problem to solve than denial. If someone wants to move but feels stuck, that's a solvable problem. So I am encouraged at our ability to collectively move forward on harnessing the full power of AI.
Here are three forms of paralysis I see regularly, and how we can address them.
1. "We don't know where to start."
My team hears this flavor of paralysis consistently in our work with CPA firms. Firms know AI is important. They feel the urgency. But the path forward seems fuzzy.
A journey of a thousand miles begins with the first one. The key is to take a first step that is small, concrete, pragmatic and safe. For example, a very common starting place for my team is onboarding Microsoft 365 Copilot for CPA firms.
But here's the catch — throwing an AI tool into the firm without context or support is a recipe for failure. That's why it's important to build the scaffolding for success. That means hands-on, experiential training and a foundational learning curriculum, even for tasks as seemingly simple as redrafting an email. It's about enabling your team to learn in low-risk, high-impact ways and gain familiarity, confidence and a visceral understanding of how AI works (and where it falls completely flat…anyone that has argued with M365 Copilot on how that client meeting does exist knows what I'm talking about.)
This then serves as a basis for taking the second step, third step and so on.
2. "There are too many vendors — what if we pick the wrong one and get stuck?"
This paralysis is driven by fear of making the wrong choice. This is a very valid concern — just look at how the vendor landscape has exploded. And accountants are brilliant at what they do, but you are not engineers and product managers, nor are you expected to be.
This is an area where some very strategic help has an outsized impact, and the great news is that there are more resources and support than ever to help you. Of course, there are your associations and alliances. I see this topic being very actively discussed within these trusted networks that you already belong to.
In addition, there are a growing number of excellent consultants entering the profession. Especially with the rise of private equity and reallocation of resources, there are an increasing number of leaders who are now serving our profession by sharing their battle-tested knowledge on setting up scalable technical systems and vetting vendors. I am part of a network of such fractional executives, including CIOs, called CPA Consultants Collective, where we are building collective knowledge on the fast-changing accounting technology vendor landscape to help our clients.
3. "The ROI is unclear and we don't want to waste money."
Ah, the classic accountant's dilemma. If we can't tie it to a P&L line, we hesitate to spend.
When it comes to ROI, I encourage firms to embrace a robust framework that segments hard ROI and soft ROI, and quantitative and qualitative ROI.
Hard ROI refers to when you take an action and immediately see direct cost savings or revenue increases. A common example in the profession is outsourcing. From a margin perspective, once you outsource, there's a direct, immediate impact to your bottom line.
Soft ROI, in contrast, is where other actions need to happen to unlock the full value of the investment. AI is a great example. When it saves time, the time saved doesn't automatically convert to better business outcomes unless you do something with that time. That might mean readjusting headcount or building a cross-selling motion to monetize newly unlocked capacity.
Quantitative ROI is what most firms are used to tracking — it includes clear metrics like hours saved, increased billings or reduced costs.
Qualitative ROI includes less tangible, but equally critical, benefits such as improved employee satisfaction, faster onboarding, better client experience or enhanced decision-making.
The firms that recognize these nuances are the ones best positioned to turn efficiency into impact and scale.
The path forward
Paralysis is frustrating, but it's also a signal. It means firms want to move. That's a gift. So what now? Stop waiting for perfect clarity. Start with what you can adopt today, even if it's small. Use experimentation as a strategy. Redefine ROI to capture benefits that may fall outside of narrower definitions.
It's time to get moving.
