Commercial Pricing & Margin Optimizer

What it answers

The first screen answers one question: is this deal commercially defensible. Everything underneath is the working.

Contract model recommendation and comparison

Twenty structures, scored against this engagement's own evidence, with the leading alternatives explained in your numbers rather than in general terms.

Risk-adjusted cost and margin

A cost distribution from fifty thousand seeded trials, not a number with a buffer added to it. P50, P80 and P90, and the margin each one produces.

Probability of protecting target margin

Counted across the simulated outcomes rather than inferred from the expected case, so it is a measurement and not a restatement.

Negotiation corridor

The floor that protects your target, the floor that protects your minimum, and exactly how much concession sits between them and the quote.

Cash exposure

Time-phased outflow, invoicing and receipts. Payment terms move when money arrives and change gross margin by nothing, and the product says so.

Transparent calculation methodology

Every formula published, every assumption named, every version stamped. A commercial director can reconstruct the price without the application in front of them.

It prices what you actually have

Delivery evidence arrives as person-days, hours, FTE over a period, story points, percentage allocations, unit volumes, a known cost, a budget, or a mixture of them. Each is normalised by its own rule before any commercial arithmetic runs. Nothing is forced into a single unit, and no conversion is invented to make the shapes match.

If you have already built a Project Effort & Timeline Estimate, the commercial model inherits the delivery evidence, the schedule and the confidence figures, and records the exact estimate version it was built on. Revising the estimate afterwards does not reprice an approved commercial plan; it offers you a new version.

What it refuses to do

A quote built on an invented number cannot survive being asked where the number came from. Where the evidence cannot reach a currency value, the product says so and names the minimum that would change it.

• Story points with no velocity and no team cost. There is no constant that turns a point into a day.
• Percentage allocations with no absolute anchor. A percentage is a shape, not a size.
• A client budget treated as a delivery cost. Budget is what they will spend, not what it costs.
• Invented rates. KnownShift holds no reference day rate for a backend developer, and will not pretend to.
• An industry-standard margin. There is not one, and a tool that invents one tells you your own policy is wrong.

Where AI is, and is not

The commercial calculation makes 0 external model provider calls. Internal cost rates, client bill rates, discounts, client budgets, margin, quotes, negotiation floors, risk exposure and cash exposure are computed on KnownShift servers by a deterministic and statistical engine, and are not sent to a language model. That is enforced by a test that reads the engine source on every commit, not by a policy somebody has to remember.