Set up a fixed-price engagement so overruns surface early enough to act on.
On a fixed-price engagement you carry the scope risk, so the entire job of the system is to tell you about an overrun while you can still act on it. Judge your setup by that standard: will this configuration warn me in week five rather than at delivery?
Contract value. The agreed price. This is the ceiling on revenue for the engagement, and everything downstream is measured against it.
Cost basis. Which people are on the engagement and what they cost you. Margin is meaningless without this. If cost rates are missing or stale, the engagement will look profitable right up until it is not.
Milestones tied to scope, not to dates. This is the setup decision that determines whether overrun detection works at all.
A milestone called Sprint 4 complete tells you nothing about progress. A milestone called Checkout flow accepted by client tells you exactly how much of the promised outcome exists. Date-based milestones always look on track until they suddenly do not, because time passes on schedule regardless of what got built.
Budget burn against scope completion.
Spend alone is not a signal. Sixty percent of budget consumed is fine at 60 percent complete and a serious problem at 30 percent complete. The ratio is the signal, and it is the reason milestones have to represent scope rather than calendar.
Review it weekly. A reasonable working rule: treat any gap over 15 percentage points between budget consumed and scope delivered as a conversation with the client rather than something to absorb quietly.
The most common way a fixed-price engagement loses money is a series of accommodations too small to feel worth a change order.
The discipline that works is logging every scope change at the moment it is requested, including the ones you fully intend to absorb for free. You are not obligated to bill for all of them. You are obligated to know what they cost, because the fourth small favor is only visible as a pattern if the first three were written down.
This also changes the conversation with the client. Saying no to the fifth request is difficult in isolation and straightforward when you can show the four you already said yes to.
At close, compare final margin against the margin you assumed when you priced the work. The gap is your estimation error for this type of project, and it is the most valuable number your firm produces.
Three or four of them and you can price the next one from evidence instead of instinct. Most firms never capture this, which is why their estimates do not improve over a decade of doing the same kind of work.
Tell us what you need. We will point you to the answer or write the article.
