Audit Friendly MODERN ACCOUNTING
BUILD WITH AI

BUILD WITH AI · PART 5 OF 5

How to use an AI assistant to prep client advisory meetings faster

 

The manual part - pulling numbers, spotting trends, drafting talking points - is exactly what a general-purpose assistant handles well if you prompt it correctly.

Client advisory prep has a specific inefficiency: the accountant already knows where to look, but the act of pulling figures, comparing them to prior periods, and drafting a coherent narrative still takes an hour or two per client. A general-purpose AI assistant does not replace the judgment call at the end of that process, but it absorbs the drafting and pattern-spotting steps that consume most of that time.

BUILD WITH AI

Client advisory prep, step by step

This workflow turns a client's period-end financial export into a structured advisory agenda draft in under 30 minutes. The assistant handles the pattern-spotting and drafting; you handle the judgment and the number check.

1
 Export the client's current-period P&L and balance sheet as a CSV or copy the table directly from your accounting system - no PDF, no reformatting yet.
2
 Open a second tab with the prior-period equivalent so you have both periods ready to paste side by side.
3
 Paste both tables into your AI assistant with a clear label: 'Current period:' above one block and 'Prior period:' above the other.
4
 Send this prompt exactly, filling in the bracketed fields: 'You are helping an accountant prepare for a client advisory meeting. The client is a [industry type] business.
5
 Review the assistant's variance list against both source tables - confirm each figure and direction is accurate before reading any further output.
6
 Use the draft talking points as a starting structure; edit for client context, add any open items from your notes, and remove anything the assistant flagged that you already know has an explanation.
7
 Add a short 'items to follow up' section manually based on your own knowledge of the client's situation - this is the part the assistant cannot supply.
8
 Save the finished document as the pre-meeting agenda and share it internally or use it as your own reference notes.
The rule: A human must verify every cited figure against the source export before the document leaves your desk.

The failure mode here is almost always in the input, not the output. An assistant given clean, structured data - an exported P&L, a balance sheet snapshot, a list of open items - produces usable first-draft commentary. An assistant given an unformatted PDF dump or a verbal description of the numbers produces confident-sounding nonsense. The workflow below is built around getting the input right before writing a single prompt.

Verification is non-negotiable because the assistant will sometimes hallucinate a trend that is not in the data, or state a percentage change that does not match the source figures. The check is mechanical: every figure that appears in the final talking points document gets traced back to the export before the meeting. That trace takes five minutes and is the only step in this workflow a human cannot hand off.

The pattern is worth building into the weekly routine. When prep is structured the same way for every client, the advisor walks in having already thought through the questions the data raises - which is the whole point of advisory work - regardless of how compressed the week was.

WORKED EXAMPLE

In practice

A bookkeeping client in professional services has just closed the month. The accountant has exported the current-month P&L and the prior-month P&L as two separate copied tables, ready to paste.

THE PROMPT
You are helping an accountant prepare for a client advisory meeting. The client is a professional services business. Review the two financial periods I am pasting below and do three things: (1) list the five largest variances by dollar amount and note the direction of each, (2) identify any line item where the current-period figure is more than 15 percent different from the prior period, and (3) draft three to five talking points an advisor could raise with the client, framed as observations, not conclusions. Do not invent figures; cite only numbers present in the data I have provided. Paste both period tables immediately after this prompt before sending.

What came back. The assistant returned a ranked variance list with five items, flagged subcontractor costs as up more than 15 percent, and produced four talking points including one about gross margin compression. One talking point cited a specific payroll figure that was actually the total compensation line, not payroll alone - the label in the source data was ambiguous and the assistant picked the wrong interpretation.

How it was checked. The accountant traced each figure in the talking points back to the corresponding row in the exported tables and corrected the payroll reference before the meeting.

A constructed example. The prompt is usable as written; the figures show the shape of a result, not a measured one.

WHEN TO USE IT

 
You have clean period-end exports and a recurring advisory client to prep for.
 
Multiple client meetings fall on the same week and prep time is compressed.
 
You want consistent depth of variance analysis across all advisory clients.
 
A less experienced team member is leading the prep and needs a structured starting point.

WHEN NOT TO

 
Client data cannot be pasted into a general-purpose tool under your firm's policy.
 
The financials are too messy or mid-period to produce reliable variance comparisons.
 
The advisory topic requires knowledge of oral conversations or context not in the data.
The pitfall: The assistant states a trend or percentage that is not actually supported by the numbers provided. You notice it only when a client questions the figure in the meeting.

WHAT TO TAKE FROM THIS

 
 Verify every figure the assistant cites against the source export before any client sees it.
 
 Structured data inputs - clean CSV or copied table - produce far more reliable output than narrative descriptions.
 
 Scope the prompt to one client and one period; broad prompts produce generic commentary that needs heavy rewriting.
THE PROMPT TEMPLATE
Get next week's workflow in your inbox
Tuesdays: one Build with AI workflow with the prompt, the steps and the verification rule. Thursdays: the explainer and the week's tools.

SPONSORED

Gusto: full-service payroll that posts clean journal entries to your ledger
Federal, state and local filing in all 50 states, native two-way sync to QuickBooks Online and Xero, and published per-employee pricing with no contract. Audit Friendly scored it 74/100 in a fact-checked review; we earn a referral fee if you sign up, and the score is not affected.

QUESTIONS THIS ANSWERS

Can I use this workflow with any AI assistant or does it require accounting-specific software?

The workflow uses a general-purpose assistant and works with any tool that accepts pasted or uploaded structured data. No accounting-specific AI platform is required, though purpose-built tools may add controls for data privacy.

What file formats work best when pasting client data into an AI assistant?

A copied table from a spreadsheet or a plain-text CSV export gives the assistant clean structure to work from. Unformatted PDF exports and verbal descriptions of numbers are the most common causes of unreliable output.

How do I handle confidential client data when using a general-purpose AI assistant?

Check your firm's data policy and the assistant's terms before pasting identifiable client figures. Many firms anonymize or aggregate data before using a general-purpose tool, then re-apply client-specific context when reviewing the output.

SOURCES

Where this comes from

Upload your resume once →

The career portal does the rest.

GO DEEPER

Go deeper

Software reviews → Fact-checked vendor reviews with verdicts and pricing
AF Stack Designer → Design your accounting stack in ten minutes
Skills that pay → Which named tools move the number
Software in postings → Every tool ranked by live demand
Accounting and finance job board → The Audit Friendly job board, every posting verified live

IN THIS SERIES

Previously: Build a reviewer-ready workpaper with AI

Next: Build with AI: client document extraction (coming)

An original workflow written for practitioners. Replicate it in a sandbox first; nothing here replaces your review.

How accountants are using AI, automation and smarter workflows to close faster, audit cleaner, and free up time for real work.

Accounting Stack · Audit Friendly Data · Accounting & Finance Jobs

Audit Friendly · modernaccounting.ai

© 2026 Modern Accounting.
beehiivPowered by beehiiv