Client advisory services: where the delivery hours go

Most guides to client advisory services cover the same ground: what CAS is, why firms should offer it, how to package and price it. What they rarely show is where the hours go once an engagement is running — what it takes, month to month, to keep one advisory client well served between calls.

This article looks at that part of the work. Specifically, it examines the preparation and continuity work inside a CAS engagement: what has to be true before an advisory conversation is worth having, and what has to be carried between conversations. It does not estimate the full cost of delivering the engagement. The conversation itself, the underlying bookkeeping, and the firm’s other obligations sit outside it.

We do not have a measured figure for this work yet, and we have not found one published that we would rely on. The one worked example we found online exists to make a case for offshoring, so we have left it out. What we can offer is a breakdown of the work, one range that firms report to us, a way to measure the rest without double-counting, and a distinction most guides skip — between the hours spent rebuilding a client’s picture and the hours spent using it.

Why these hours are hard to see

The work is easy to miss because it does not sit in either place a firm already measures.

Bookkeeping looks backward. It closes the month, reconciles the accounts, produces statements. Firms know what that costs and have priced it for decades.

CFO-level work looks forward. Forecasts, scenarios, a fundraise, a board pack. Firms know what that costs too, because it is scoped as a project or billed as a fractional retainer.

Advisory continuity lives in between. It is the work of knowing what changed in a client’s business since last month, what was discussed last time, what was decided, and what the client said they would do. In many firms that work is not a line item. It is held by whoever runs the relationship, rebuilt before each call, and not recorded in a form the firm can reuse.

Four kinds of work between calls

Take one advisory client on a monthly cadence. Between one call and the next, the firm does four kinds of work whether or not it tracks them. Each is worth measuring separately.

Preparing for the call. Pulling current numbers, comparing them to last month, working out what moved and why, deciding what to raise. Without a running record, this is reconstruction: reading back through the books and the last set of notes to rebuild the picture.

This is the one category we can put a range on. In conversations with CPA firms, preparation for a single advisory call runs from around two hours to half a day, depending on client complexity. That is what firms report to us, not something we have measured, and the spread is wide. But it recurs often enough that we are comfortable repeating it.

Answering questions between calls. “Can I afford to hire?” “Why is cash lower than you said?” “What happens if the big invoice slips?” Each one takes someone back into the books. Some firms bill these; many treat them as part of the relationship and do not. Either way, the time is real and usually uncounted.

Keeping the forward view current. An advisory client has two forward numbers. The budget is the plan the client agreed to; it stays fixed as the reference. The forecast is the current expectation, and it moves as new information arrives — a delayed receivable, a hire brought forward, a slower month. Someone has to update the forecast against what the books and bank now show, and explain the variance to budget. Left alone, the next call is built on a number nobody trusts.

Carrying what was agreed. Last month the client said they would chase the overdue receivable, delay the equipment purchase, and send the new lease. This month someone has to remember all three, check which happened, and raise the ones that did not. If nobody does, the call restarts from zero.

Measuring it without double-counting

If you want your own figure, the categories above are the buckets. Three rules keep the count honest.

Log each activity against the client, the person doing it, and one work category, and count each block of time once. Recording the person matters because capacity is calculated per role: a senior’s ceiling uses that senior’s hours per client, not partner and staff hours combined. Count preparation only up to the start of the call; the call itself is a separate cost. And keep between-call questions distinct from preparation, even when a question does the prep for you: if answering “can I afford to hire” means updating the forecast, that time goes in one bucket, not both.

Two or three months of this across a few clients will give you a per-client range. It will vary with complexity, which is fine — the point is to see the shape, and to notice how much of it is rebuilding something the firm already knew a month ago.

What tends to happen as the engagement ages

This part is observation rather than measurement, so treat it that way.

In the first month there is no history to carry. Over the following months the engagement accumulates one: previous conversations, open commitments, a forecast that has moved several times, and a client who expects the firm to remember all of it. Firms tell us the strain often shows around the third month — not because anything is special about the number three, but because that is roughly when the accumulated context exceeds what a person can reliably hold in their head and rebuild from notes.

At that point the firm makes a choice, usually without naming it. Keep carrying the thread by hand, which costs hours. Or let it thin out, in which case the calls get shorter, the client asks fewer questions, and the engagement drifts back toward bookkeeping with a monthly meeting attached.

Two different limits on an advisory book

Guides tend to treat advisory capacity as one number: how many clients a person can carry. It is more useful to think of two limits and ask which one you are hitting.

The reconstruction limit. If most of the between-call work is rebuilding each client’s picture from scratch, then capacity is bounded by that rebuilding time. Divide the hours a person has free for advisory by the per-client hours across the four categories, and you have a ceiling.

An illustrative case, using only the reported preparation range: a senior with roughly a day and a half a week free for advisory has around 48 hours a month. At two to four preparation hours per client, preparation alone would support somewhere between 12 and 24 clients — before questions, forecast updates, and commitment tracking are added. Those other three categories will bring the real number down, possibly by a lot, and this calculation makes no claim about what it is. Its purpose is to show that the answer depends on the rebuilding hours, and that those are the hours a firm can most directly change.

The judgment limit. Suppose the rebuilding were largely gone — each client’s picture already assembled, the forecast already updated, the open items already listed, before anyone sat down. What limits the book then is different: how many advisory conversations a person can have well, and how many client situations they can hold in their head and make good decisions about. That is a limit on judgment and attention, not on hours in the books.

How much capacity you gain by moving from the first limit toward the second depends on what delivery work remains, how much effort it takes to keep the running record current, how often exceptions need a human back in the books, and how much staff time is actually free. It will not be the same for every firm. But the direction is consistent: reducing repeated reconstruction frees time for analysis and for the conversation, and the conversation is the part the client is paying for.

This is the same shift that other fields are going through as tools take on the mechanical part of the work. When producing the draft stops being the main cost, the practitioner’s limit becomes how fast they can decide and how much context they can keep straight. Advisory is heading the same way.

For related arithmetic, see how many clients one controller can carry and what a quick question costs your firm. For preparation on its own, see how long client call preparation takes.

Three ways to reduce the reconstruction

These are not mutually exclusive, and firms often combine them.

Add a person to own continuity. A controller or senior accountant whose job is the in-between work across the advisory book. This works, and a good hire will also build the reusable records that reduce rebuilding for everyone. The constraint is cost against advisory revenue: smaller firms tell us the salary is hard to justify until the advisory book is already sizeable, which creates a sequencing problem — the capacity is needed to grow the book that would fund the capacity.

Keep it with the relationship owner. The most common arrangement, and often the right one early on. The partner or senior carries the thread, does the preparation, fields the questions. It has no hiring cost, but the hours are real: they come out of time that could go to other clients or other work, and they scale only as far as one person’s memory does.

Build the running record. Keep a per-client record — what changed in the numbers, what was discussed, what was decided, what is still owed — maintained between calls rather than rebuilt before them. When it exists, preparation shifts from reconstruction to review. The forecast is updated as the books move rather than in a rush before the call. Commitments carry forward. Between-call questions can be answered from the record more often than from the books.

A firm can do this with discipline and a shared document; it depends on someone keeping it current, and it lapses when they are busy. AgentLink is built to keep that record for each client automatically — watching the books and bank, updating the forecast against them, and holding decisions and open items — so the picture is ready before the call. It does not remove the need for a person to review it and make the calls. What it reduces is the repeated rebuilding.

What to do with this

Count where the month goes for a few advisory clients, using the four categories and the three rules above. Then look at the split. If most of the hours are reconstruction — rebuilding a picture the firm already had — that is the part worth attacking first, because it is the part that can be removed rather than just paid for. How much capacity that frees is something each firm has to test with its own numbers. We will add a measured example here as soon as we have one worth publishing.