Client Profitability at a Consulting Firm: Which Engagement Types Actually Carry Margin

The problem: The work, the hours, and the billing sit in separate systems, so nobody can say which client types and engagement types actually carry the firm's margin.

The solution: Bring hours, scope, and realized billing together against each engagement, so profitability is a number per client type rather than a partner's impression.

Most managing partners can name the firm's biggest clients from memory and can name the ones that are difficult to work with. Almost none can say which kinds of clients and which kinds of engagements actually make money. Those are different lists, and the gap between them is where the margin goes.

It is a fair question and it sounds answerable. Which pays better: the long retained advisory relationships or the short diagnostic projects? Do fixed-fee engagements beat time-and-materials once the overruns are counted? Is the mid-market client, who takes less handholding, worth more per partner hour than the large account everyone is proud of? A firm doing $8 million a year has years of evidence on all of it. It just cannot get to it.

Why the answer will not assemble

The reason is boring and structural. The three facts you need to answer the question are recorded in three different systems by three different people at three different times.

The scope lives with the engagement: what was sold, at what fee, under what arrangement. The hours live in a time system, entered by consultants, and they exist mainly so invoices can be produced. The billing lives in accounting, and it records what went out the door and what came back, which is not the same thing, because write-offs, discounts, and unbilled overruns all happen after the timesheet is closed.

Any one of those systems can produce a clean report about itself. Time can tell you utilization. Accounting can tell you revenue by client. The engagement record can tell you what was promised. None of them can tell you realized margin by engagement type, because that requires the three sitting in the same row, and nothing puts them there.

So a partner asks for it, someone spends two weeks exporting and reconciling, the answer comes back with caveats, and the decision it was meant to inform has already been made. The firm goes on selling whatever it sold last year.

What you cannot see, specifically

The gap has a shape. It means the firm cannot answer:

• Which client types return the most margin per partner hour, which is the scarcest thing the firm sells.
• Whether fixed-fee work is genuinely more profitable, or whether the overruns absorbed after the fee was set have quietly made it the worst category.
• What the realized rate is by engagement type, as opposed to the standard rate on the rate card, which is a number nobody actually collects.
• Which clients consume unbilled time between engagements, in relationship management, scoping that goes nowhere, and revisions never invoiced.
• Which practice areas are subsidizing which, and by how much.

The last one is the expensive one. In a firm at this size there is almost always one category of work everyone believes in that is running below the firm's average margin, and one nobody talks about that is quietly carrying it. Without the numbers, the belief wins, and the firm keeps steering people and business development toward the work that feels most like the firm.

How the answer becomes available

The fix is not a better spreadsheet and it is not a new time system. It is joining what already exists.

Every engagement gets a record that carries its type, its client type, and its fee arrangement. Hours land against that engagement rather than only against a client. Realized billing, after write-offs and discounts, comes back to that same engagement rather than stopping in accounting. Once the scope, the hours, and the money actually collected sit together in one connected picture of the firm, margin by engagement type stops being an exercise and becomes something the picture already knows.

Then the repetitive part comes off people. Nobody rebuilds a profitability model each quarter. Automation watches engagements as they run and raises the exceptions: the fixed-fee job that has passed its budgeted hours in week three, the client whose realized rate has slipped two quarters running, the engagement type whose write-offs have crept up while its billings held steady. The finance lead stops assembling the picture and starts acting on it, before the engagement closes rather than a quarter after.

A look at a consulting firm

Consider a management consulting firm doing about $8 million a year with roughly 40 people, split across a few practice areas. Some work is retained advisory. Some is fixed-fee project work. Some is time-and-materials. Clients range from mid-market companies to a handful of large accounts that dominate the conversation internally. The firm is profitable and growing, and the partners' shared sense of which work pays best has never been tested against a number.

Suppose the firm connects engagement scope, logged hours, and realized billing into one picture. Within two quarters you would expect the book to sort into tiers that surprise people.

Put a number on it. If roughly $2.4 million of the $8 million, meaning around 30 percent of billings, sits in engagement types running about ten points below the firm's average margin, that is on the order of $240k a year of margin the firm is giving up without a decision ever having been made to give it up. In a firm this size, that is comparable to two senior hires.

What the partnership would likely do with that is not dramatic. Some fixed fees get repriced with the real hour history behind them, which is a much easier conversation with a client than an intuition is. One engagement type gets restructured so the overruns land somewhere other than the firm's margin. Business development shifts toward the client type with the best return per partner hour, which is often not the largest logo. And the debate about which practice area matters most stops being a debate, because it becomes a table anyone can open.

How to start

You can begin with one quarter of history and the systems you already have.

1. Give every engagement a type. Client type, engagement type, fee arrangement. Three fields. Without them, no amount of data produces the answer.
2. Get hours onto the engagement. Not just onto the client. The engagement is the unit the pricing decision gets made at.
3. Bring back what was actually collected. Set realized billing, after write-offs and discounts, beside the hours. The difference between billed and realized is where most of the missing margin lives.
4. Rank, do not average. Sort engagement types by margin per partner hour rather than by revenue. The ranking is the decision.

The takeaway

An $8 million firm does not have a demand problem, it has a blindness about which demand is worth taking. The evidence for which client types and engagement types carry your margin is already in your time entries, your engagement records, and your billing history. It has never been assembled because those three live apart, and by the time someone assembles it by hand the answer is stale. Put the scope, the hours, and the money actually collected in the same row, rank the engagement types by margin per partner hour, and the next thing you sell is likely to be a different thing than the last.