Skip navigation
Advisors and AI
open-ai-logo-eye.jpg JOEL SAGET/AFP/Getty Images

Can ChatGPT Help Fix My Practice?

We posed a hypothetical practice management dilemma facing an advisory firm to ChatGPT. Then we asked three experts to weigh in on the merits of the advice.

Much of the fuss about OpenAI’s ChatGPT has centered on whether it can offer solid advice for clients; but what if advisors turned the program back on themselves? How does ChatGPT do when giving practice management advice to financial advisors? We posed a hypothetical dilemma facing a hypothetical advisory firm. Then we asked three practice management experts to weigh in on the merits of the advice.

Here’s a look at the situation and ChatGPT’s (edited) guidance, plus the experts’ assessments.

The dilemma:

An advisor runs a 20-year-old RIA with around $2 billion in assets under management. For a variety of reasons, over the past two years, four employees have left the firm—two advisors and two client service employees. (The reasons for their leaving included a career change, moving to another part of the country and needing to take care of an elderly parent). The advisors who left were experienced, with at least seven years at the practice. Now the firm’s owner, age 50, is having trouble replacing the departed employees—particularly with highly experienced candidates.

What should he do?

ChatGPT’s advice:

The financial advisor should take several steps to address the problem:

  1. Analyze the reasons why employees have left. The advisor should conduct exit interviews or surveys to identify any patterns or issues that may be causing employees to leave;
  2. The advisor should review the compensation and benefits package to ensure it is competitive with other firms in the industry;
  3. The advisor should consider expanding the recruitment pool to include candidates from different geographic locations or backgrounds;
  4. Invest in training and development programs to help new employees get up to speed quickly and develop the skills they need to be successful in their roles. This can also improve employee retention; and
  5. The advisor may want to consider outsourcing certain functions or partnering with other firms to help fill any gaps in expertise or experience.

The experts:

David DeVoe, founder and CEO, DeVoe & Co., categorized the advice as a highly simplistic assessment of a complicated situation. “The CEO of a $2 billion AUM firm would know everything in that document before they had their morning cup of coffee,” he says.

The big weakness: a lack of industry knowledge. “This is advice that could apply to any industry,” he says. “There’s a lack of real functional expertise: We’re talking about complex problems here.” What the client needs, instead, are questions that stem from an in-depth understanding of the sector and the issues and problems that expertise might uncover.

For example, there’s the matter of assessing just how bad the turnover is. “A $2 billion firm probably has about 20 employees, and a 10% attrition a year is an unusually high number,” he says. “But ChatGPT wouldn’t know the number of employees a $2 billion firm might have or even what the attrition rate in the industry is.”

Other recommendations only touch the surface of what needs to be evaluated. For example, according to DeVoe, while assessing compensation and benefits is a good idea, it’s only one of multiple issues to study. “We need to know, what’s your capacity? Are your people feeling overwhelmed? Are there any recent events that could have driven the attrition?” he says.

Devoe liked a few points, however, like the AI’s recommendation about taking steps to determine whether the firm faces internal challenges. Ultimately, “It’s an interesting tool and may change the industry,” he says. “But we’re at the first inning here.”

Bill Van Law, CEO and founder, WVL Group, also gave mixed reviews. “While broadly there were good ideas, they weren’t as deep as they need to be to be effective,” says Law. “And the advice doesn’t address the most important aspect, which is implementation.”

For example, he agreed it’s important to use exit interviews and surveys to analyze why employees have left. That’s particularly true because high turnover could be a sign of underlying problems likely to affect client service, retention and growth. But those steps should be just the beginning, according to Van Law. “You need to dig deeper with remaining team members, finding a way to get candid feedback without a fear of potential repercussions.”

Or take the issue of compensation and benefits. They’re key to remaining competitive, but they’re also only part of the story. “It’s often the nonfinancial factors that can really lead to turnover,” he says. Assessing those elements requires a much larger analysis of the work environment. Related to that is the matter of training and development. For Van Law, the AI’s advice is not so much off the topic, as highly inadequate. “It’s the very tip of the iceberg—one really small element to consider,” he says. The firm needs to take a variety of other steps, all aimed at providing employees with a career path.

Marty Miller, partner, Strategy & Resources, regarded the advice as surprisingly helpful, at least as suggestions that can serve to get you started. “It moves you nearer the 50-yard line,” she says. Some of the better recommendations: surveying employees, making sure there’s a competitive compensation plan, expanding the recruiting pool geographically (depending on the role) and more training and development.

Still, she also agrees with DeVoe and Van Law. “It doesn’t replace wisdom and lacks industry nuance,”  she says. Meaning: The answers are pretty generic. “If the situation were about a $2 billion in AUM firm or a $100 million firm, it would have come up with the same answer,” she says. But the questions you’d ask of that larger practice should be substantially different from the ones you’d raise with the smaller one.  “This is one size fits all,” she says.

TAGS: Technology
Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.