The 10-minute advisory prep workflow accountants are using with AI

All challenging tasks are won in the preparation stage. You can’t just wake up one day and run a marathon. Well, you technically can but it’ll be more painful than it needs to be. 

The training, the energy management, the mental composure. This is all built and honed during the months leading up to the big race. Getting the preparation right is the difference between not finishing and a PB.

The same concept is true for your advisory client meetings. 

If you’re reviewing the numbers for the first time as the client is sitting across from you, or working from a quick glance at last quarter's P&L while trying to listen to what they're telling you, you’re setting yourself up for failure. You’ll catch the obvious things, but you’ll miss the subtle ones all while spending the conversation reacting rather than guiding.

This is purely a time problem. Deep preparation for every advisory meeting takes time that most accountants don't have. So the prep gets compromised, and the quality of the conversation suffers. 

But what if you gained time back? What if you actually had the capacity to do the deep-dive your clients need and deserve? That you need and deserve? 

You probably know what’s coming next.

Yes, this is an AI opportunity. Here’s why.

What good preparation actually looks like

A genuinely well-prepared advisory meeting starts with knowing three things: what's happened in the client's financials since you last spoke, what it means, and what you should do about it.

Good prep identifies the trends that matter, including what’s driving variances and if they’re structural or one-offs. It flags the risks the client may not have noticed, and it prepares you to ask the questions that move the conversation forward.

Painting that picture from scratch can take 30 to 45 minutes per client. Multiplied across a week of advisory meetings, that's a significant chunk of time. And that’s why so many accountants end up doing lighter prep than they'd like.

So what’s the alternative?

In a live Claude Training for Accountants session, Twyla Verhelst, CPA and Bryon Patrick, CPA demonstrated how AI can help augment the prep process, potentially reducing the time by a third of the manual process. 

Using a fictional accounting firm, they show how Ryder from Cedar & Vine uses AI to prep for an advisory meeting with a client.

The AI-assisted advisory prep workflow

Here’s a scenario with you in the driver’s seat: You’re two days out from a client’s quarterly review. You open Claude, share the client's recent financial statements, and give it a prompt like this:

I'm meeting with [client name] on Thursday for our quarterly advisory review. Here are their financials for the last three periods. Please:

1. Identify the most significant trends or changes in their financial position

2. Flag any areas of concern I should raise with them

3. Suggest three to five questions worth exploring in this conversation

4. Draft a short pre-meeting brief I can review before the call

Within a few minutes, you have a structured brief:

• It's identified that gross margins have compressed by four points over the past two quarters.

• It's flagged that receivables days have crept up to 52 days—a pattern that could suggest a cash flow issue in the next quarter. 

• It's noted that payroll costs increased faster than revenue. 

• And it's drafted three questions to ask your client: what's driving the margin compression, whether any large receivables are at risk, and what the business is planning for Q4 given the trajectory.

You review the brief, add your own context—you know this client just hired two new people, which explains the payroll increase—and you walk into the meeting already thinking at the level of the conversation you want to have.

The whole process, from opening Claude to having a structured brief in hand, takes under 10 minutes.

Twyla said it herself, the potential gains AI has for advisory firms is an exciting prospect. “I can't help myself when it comes to advisory,” she confessed during the webinar. “I ran an advisory firm and I wish I had AI at the time.”

What AI doesn't replace

The brief is a starting point. What you do with it is the job.

The insight that those two new hires are actually the right strategic move for where this client is heading—that's yours.

The judgment about whether the margin compression is a pricing problem or a cost problem, and which conversation to have first—that's yours.

The relationship that means this client will actually listen when you raise the cash flow concern rather than getting defensive—yep, that's yours too.

This is what the most valuable advisory work has always been: the interpretation, the professional judgment, the relationship that gives the conversation weight.

AI doesn't replace any of that. What it replaces is the data work that happens before you get there. AI can handle the reading, the pattern recognition, and the structuring of what you already know into something ready to act on.

When you arrive at a meeting having worked through the client's numbers beforehand (rather than scanning them for the first time as they sit across from you) the conversation changes. Clients notice your level of detail. They’ll ask more questions, and they’ll trust your guidance more. The meeting becomes worth more to them, which is ultimately what determines whether they think of you as their accountant or their advisor.

In Twyla’s demo of this workflow above, she uses AI to enhance the human layer of the client relationship. She prompted Claude with “ What's one thing I should personally remember to do on the start of tomorrow's call?" 

And because, in this hypothetical scenario, the fictional accountant, Ryder, fed Claude with transcripts from previous meetings, it identified a snippet from the most recent call where the client was discussing their daughter’s pageant.

That’s what Twyla calls “a moment.”

“ On every call you have with a client,” she explains, “there's at least one golden nugget that is an element of personal information that you can use to build a relationship.”

In fact, according to The State of AI in Accounting Report 2026, 48% of accounting professionals state that AI is benefitting them by enhancing client relationships and communication. Al is transforming how accounting professionals work and how they connect.

Twyla’s example is a solid and simple instance of using AI to make your client relationships more personal and intentional. And all it took was the right data and the right prompt.

Building AI prep into your practice with Claude

If you’re using Claude, the way to make this consistent and easily repeatable is to build it as a Claude skill. Claude skills are saved instruction sets that run the same process every time. 

In this example, your Advisory Prep Skill will run the prep process the way you’ve instructed it to, using the specific context and framing you want for your practice. You do this once, refine it over a few uses, and then it runs reliably.

The starting prompt above is a good foundation. Customize it to include the metrics you care about for a given client type, the concerns that are specific to their industry, and the format that works for how you prepare. Then save it. If you want a more in-depth foundation, revisit Twyla’s video above for some inspiration.

For firms thinking about how to structure repeatable advisory services, this is a useful place to start. Identify the one or two client types you serve most often, build a prep skill for each, and let consistency do the rest.

Just like interval training helps build strength for those marathon hill climbs, this practice will help strengthen your advisory muscle, one AI-augmented prep at a time.