What one advisory client is actually worth

Short answer: a mid-tier advisory client on a $4,000 monthly fee is worth about $48,000 a year in revenue and, if the deliverable is built by hand, somewhere around $20,000 to $25,000 in margin after senior time. Fix the build cost and the same client throws off $38,000 or more. The fee is not the problem in most CAS practices. The cost of producing the deliverable is.

These are the numbers I ran on my own practice before I decided to build software instead of hiring another analyst. I am using round figures so you can swap in your own.

The revenue side is the easy part

Take a client in the middle band, $4,000 a month for a rolling forecast, budget versus actual and a monthly decision meeting. That is $48,000 a year. Advisory clients also stick: churn in a well-run CAS practice runs well under compliance-only churn because you are in the room for the decisions, so a three-year relationship is a conservative assumption. Call it $144,000 of lifetime revenue from one client, before any price increases and before the referral they send you in year two.

The CAS benchmark data backs this up. Median net client fees per professional came in at $156,250 in the most recent AICPA and CPA.com survey, up 29 percent, and practices with a CFO-level tier earn roughly 30 percent more per client per month than practices without one.

The cost side is where the practice actually gets decided

Here is what that same $4,000 client cost me each month when the model lived in a spreadsheet.

• Monthly refresh: pull the ledger, reconcile the model, update the drivers, rebuild the cash view. Roughly six hours of a senior person.
• Deliverable prep: formatting, the variance narrative, the deck. Another three to four hours.
• The meeting itself: one hour, plus prep. This is the only part the client would describe as advisory.

Call it eleven hours a month. At a fully loaded senior cost of $150 an hour that is $1,650, or about 41 percent of the fee, spent before anyone has given the client a single piece of advice. On a $48,000 client that is roughly $20,000 a year of cost that produces nothing the client sees.

Then add the onboarding week. Mapping the chart of accounts, building the three statements, reconciling to history, getting the first forecast to tie out. Forty hours was typical for me. That is $6,000 of senior time in month one against a $4,000 invoice. Every new client started underwater.

The margin math, side by side

• Hand-built model: $48,000 revenue, about $20,000 of refresh and prep cost, $6,000 of onboarding in year one. Year-one margin roughly $22,000. Year two, roughly $28,000.
• Model builds itself: $48,000 revenue, about $2,400 a year of software for that client file, two hours a month of review and meeting prep at $150, so $3,600. Onboarding the same afternoon. Year-one margin roughly $42,000.

Same client, same fee, same meeting. The difference is close to $20,000 a year per client, and it comes entirely from what the deliverable costs to produce. Across a book of 20 advisory clients that is $400,000 of margin a year that does not require selling a single additional engagement.

Capacity is the second dividend

The margin number is the one partners notice first. The capacity number matters more over three years. At eleven hours a client, one senior person tops out at seven or eight advisory clients before something slips. At two hours a client, the same person can carry 15 to 20 and the constraint becomes calendar, not production. The benchmark data shows the same pattern from the other direction: practices that invest in technology serve about 50 percent more clients per professional than those that do not.

That is the real reason I never made the analyst hire. Every analyst I priced out would have let me take on about six more clients and added a fixed salary I was covering whether those clients showed up or not. Software let the people I already had take on twelve more.

Run this on your own book

• Pick your three most typical advisory clients. Write down the fee.
• Have the person who prepares the deliverable log one month of real hours. Not the estimate. The log.
• Multiply by their loaded cost. That is your production cost per client.
• Compare it to the fee. If production is over 20 percent, you have a build problem, not a pricing problem.
• Do the same for the last client you onboarded. Include the week.

Questions I get on this

Doesn't the client expect to see the work? The client expects to see the answer. In five years nobody ever asked me how long the model took. They asked whether they could make the hire.

What about the junior staff who do the refresh today? They move to the review and the meeting prep, which is the part that teaches them to be advisors. The refresh never taught anybody anything.

Isn't per-client software cost just another line item? It is, at roughly five percent of a mid-tier fee. The line item it replaces was 40 percent.