What a deal actually nets after you deliver it

Most businesses know what they charge. Fewer know what they keep. And in a services business the gap between the two is usually wider than anyone in the room believes when the deal is signed.

The price is decided at one moment, by one person, using one number. Everything that erodes it happens later, gradually, in someone else's part of the business. By the time the margin is gone the deal is six weeks old and nobody connects the two.

The four things that eat a deal

1. The discount given to close it. Visible, deliberate, and the only one anybody tracks.
2. Delivery cost above what was scoped. Usually hours, usually unbilled, usually nobody's fault in particular.
3. Overhead the deal consumes but never gets charged. Project management, account handling, revisions, the calls.
4. Payment terms. Net 60 on a job you funded yourself is a cost, even though it never appears as one.

Individually each looks small. Together they routinely take a healthy-looking engagement to nearly nothing.

What that looks like with numbers

Take a $40,000 project, scoped honestly: 220 delivery hours at a loaded cost of $95 is $20,900; a subcontractor at $6,500; project management of 40 hours at a loaded $85 is $3,400; travel, software and pass-through at $1,200.

Direct cost is $32,000, so the deal carries $8,000 of contribution, or 20%. Not spectacular, but real.

Now run it through what actually happened.

A 7% discount to get it signed. Revenue drops to $37,200 and contribution falls to $5,200 — 14%. The discount was 7% of the price and 35% of the profit.

Thirty-five hours of unbilled scope. A few extra revisions, one additional stakeholder, a report nobody planned. At $95 loaded that is $3,325, and contribution drops to $1,875 — 5%.

Net 60 payment terms. At a 10.5% cost of capital, financing $37,200 for sixty days costs about $642. Contribution lands at $1,233.

A deal priced at 20% delivered 3.3%. Nothing went wrong. No one was negligent. Every step was normal.

Scope creep is a margin event, not a service issue

This is the one worth changing first, because it gets treated as a relationship question when it is an arithmetic one.

When a client asks for something outside scope, the person who receives the request is usually the person least equipped to price it and most motivated to say yes. They are protecting the relationship. Nobody has told them that thirty-five hours is a third of the profit.

Two things fix most of it. Tell delivery staff the contribution margin on their own projects — not the revenue, the margin, because people protect a number they can see. And make a change order the default response rather than the awkward exception: a line, a price, an email confirmation. Clients accept change orders routinely when they are normal, and resent them when they arrive as a surprise at invoicing.

Terms are part of the price

A client who negotiates you from net 30 to net 60 has extracted a discount without either of you calling it one. On a business running a 52-day DSO, that is real money, and it compounds across every engagement.

It also cuts the other way, which is the useful part. If terms have a price then they can be traded. A 2% early-payment discount that moves a client from day 60 to day 10 is often cheaper than the financing it replaces, and much easier to agree than a rate increase. A 30% deposit on project work does more for your cash position than most price rises, and clients object to it far less than owners expect.

Discounting, priced properly

Discount authority is almost always set against revenue, which is the wrong denominator. At a 20% contribution margin, a 10% discount removes half the profit on the job. At 45%, it removes 22%.

The rule worth adopting: express every proposed discount as a percentage of contribution before anyone approves it. "Ten percent off" and "half the margin" describe the same decision, and only one of them gets refused.

This is the arithmetic sitting underneath your break-even point, and it is why a business can grow revenue through discounting and go backwards in cash.

Then look at the portfolio, not the deal

Deal-level margin is where the fixing happens. Pattern-level margin is where the strategy is.

Run the same calculation across a quarter of completed work and group it — by service line, by client size, by who sold it, by whether it was fixed-fee or time and materials. The distribution is almost never flat, and its shape is usually a surprise:

• The largest clients often carry the worst margins, because they negotiate hardest and consume the most account management.
• Fixed-fee work either substantially beats or substantially trails time and materials. It rarely matches it.
• One service line usually subsidises another, and nobody has decided that it should.

That is a unit economics question as much as a pricing one, and it is the difference between raising prices across the board and raising them where they are actually wrong.

What to put in place

• A floor, not a target. Decide the contribution margin below which you decline the work, and hold it. Work below the floor makes the business busier and no better off.
• Post-delivery review on anything material. Compare scoped hours to actual on every engagement over a threshold. Fifteen minutes per project, and it surfaces the estimating errors that repeat.
• A standing deposit policy. Not negotiated per deal, so it never becomes a concession.
• Margin visible to the people who spend it. The highest-leverage change on this list, and it costs nothing.

Run it on your numbers

The pricing profitability calculator takes a deal price, delivery cost, overhead allocation and terms, and returns what the engagement actually nets. No email required.

Run your three largest deals from last quarter through it first. If the biggest ones are the thinnest, that is not a pricing problem — it is a negotiating-authority problem, and it gets fixed somewhere else.