Short Overview

Price the Work Before You Plan It

A short introduction to the CAPM model and the problem we're trying to solve with it: most agencies price work only after they have started imagining delivery. Learn what we're borrowing from finance, why the hybrid approach is best understood as pricing governance rather than prediction, and how proposed deal economics either clear the hurdle or fail it. We also explain why enterprise agencies may lean toward the pure approach while small and mid-sized agencies often begin with the hybrid one.

Typical flowImagine delivery, then justify price.Risk gets priced backwards.
Better flowName risk, set the hurdle, then price the work.Delivery planning follows a clearer commercial threshold.
Decision resultGo, reprice, discovery, or decline.The deal clears the hurdle or it does not.
Next stepsRead the Walkthrough, then the Decision Guide.Use the cards after you see one example and the procedure.
Start here if you want the thesis without the theory. This is the best first read for most people.

Reading order

This TL;DR is the best first read for most people. Next read the Walkthrough → Decision Guide → Decision Cards. The Calibration Notes and Theory documents are deeper dives into CAPM and how it's used here. If you want a simpler guided alternative, smaller agencies can use the Small Agency Version.

Open the Decision Cards Overview Walkthrough Decision Guide Calibration Notes Discovery First Theory
CAPM for Agencies

Agencies price risk backwards.

Open the live cards

If you want to move between the text and the tool, jump straight into the interactive decision cards:

They plan the work, estimate the hours, add some padding, and hope the margin holds. When things feel uncertain, they increase contingency. When things feel safe, they get aggressive.

It works until it doesn't.

The problem is not estimation by itself. The problem is that most agencies never separate the kinds of risk they are taking on. They treat all uncertainty the same, even though some risks are project-specific and some affect the whole business at once.

That distinction matters.

In financial economics, the Capital Asset Pricing Model exists to think clearly about exactly this problem. It separates risk into categories and asks a simple question:

What return is required for this level of exposure?

That idea translates well to agency work, especially if you treat it as a governance model for pricing decisions rather than as a literal asset-pricing engine.

Clarification: Price before you plan is narrower than it sounds: price before delivery planning is committed, not before the solutions team has done enough technical assessment to judge complexity, client concerns, and whether implementation or discovery is the thing to sell.

The Core Idea

Every client engagement is an investment.

You are investing:

  • time
  • capacity
  • reputation
  • opportunity cost

The question is not just:

Can we deliver this project?

It is:

Should we take this project at this price?

Translated to agency work, the model becomes:

Minimum Acceptable Margin = Base Margin + β × Risk Premium

Where:

  • Base Margin is what you earn from your safest work
  • Risk Premium is the extra margin your project work historically earns
  • β is how risky this engagement is relative to your average

That gives you a hurdle rate.

But the decision is not finished there.

You still have to compare that hurdle to the actual economics of the deal:

  • quoted deal price
  • estimated delivery cost
  • resulting proposed margin

That is the real go/no-go test.

Where Different Agencies Start

For enterprise and international agencies, the cleanest starting point is usually the pure approach. Their risk exposure more clearly tracks macro forces like currency shifts, cross-border regulation, global talent markets, and platform dependencies at scale.

For small and mid-sized agencies, the practical starting point is usually the hybrid approach. When you only have a handful of active deals, one difficult engagement can still behave like a portfolio event, so folding deal-specific risk into pricing governance is often the more useful move.

What Most Agencies Get Wrong

Most agencies:

  • pad hours
  • add contingency percentages
  • rely on intuition

But they do not distinguish between:

  • risk that washes out across projects
  • risk that hits everything at once

That leads to two problems.

1. Underpricing Systematic Risk

When:

  • talent markets tighten
  • platforms shift
  • budgets freeze
  • rate pressure increases

every project gets harder at the same time.

If you have not priced that in, your whole portfolio suffers.

2. Mishandling Engagement Risk

Client quirks, unclear scope, contractual structure, and technical surprises are real. They matter. But they are not the same kind of thing as market-wide exposure.

The practical mistake is not just misclassifying them. It is failing to name them consistently before a deal closes.

The Two-Layer Model

To fix this, the model separates risk into two layers.

Layer 1: Systematic Risk

This is the environment the agency operates in.

Examples:

  • platform stability
  • talent market
  • economic conditions
  • regulatory exposure
  • revenue concentration
  • rate pressure

This layer sets the systematic adjustment factor that frames the pricing environment for every deal. In the Decision Cards, that work happens in the Layer 1 card.

Layer 2: Engagement Risk

This is the specific deal.

It is a presales assessment, not a full delivery plan. The point is to get enough input from the solutions team to judge complexity, client concerns, and scope risk before commitment. If that still is not enough to price implementation responsibly, then discovery is the thing to price. In the Decision Cards, this is the Layer 2 card.

You score:

  • client track record
  • scope clarity
  • technical complexity
  • internal capacity
  • contract type
  • political complexity
  • timeline pressure

This layer gives you the engagement score and engagement beta used in the hybrid model.

Turning Risk Into A Decision

In the hybrid calculator, the current logic is:

  • Engagement β = Engagement Score / 21
  • Blended β = Engagement β × Layer 1 factor
  • Required Margin E(R) = Rf + Blended β × (Rm - Rf)

Then:

  • Proposed Margin = (Deal Price - Delivery Cost) / Deal Price

Now you can compare the proposed margin against the hurdle.

That comparison is the point.

Without it, you only know what the deal should clear in theory. With it, you know whether the quoted work actually deserves a yes.

If you are a mission-driven or certified agency, the B Corp card adds an impact adjustment that shifts the standard hurdle from E(R) to E(R*).

A Simple Example

Say:

  • base margin Rf = 10%
  • portfolio margin Rm = 22%
  • risk premium = 12%

And after scoring the deal and current environment, the model gives you:

  • Required Margin E(R) = 29%

Now assume:

  • Deal Price = $120,000
  • Estimated Delivery Cost = $90,000

That means:

  • Proposed Margin = 25%

So the decision is not:

The project feels promising.

It is:

The project is 4 margin points below the hurdle, so reprice it, reduce risk, or walk away.

That is much more useful.

Why This Is Useful

This model is not valuable because it predicts outcomes with statistical precision.

It is valuable because it enforces discipline.

It forces the agency to:

  • separate systematic and engagement risk
  • make judgment explicit
  • compare required margin with actual deal economics
  • document why a deal should proceed or stop

Most importantly, it forces:

  • sales
  • solutions
  • delivery
  • leadership

to use the same vocabulary for risk before committing.

That is the real win.

What The Hybrid Model Really Is

For most agencies, the hybrid layer should be understood as heuristic pricing governance.

It is not a statistically correct pricing engine.

Its practical utility is:

  • internal alignment
  • presales discipline
  • postmortem calibration

The math matters, but the governance effect matters more.

The Real Shift

Most agencies:

  • plan first
  • price second

This flips it:

Price the work before you plan it.

Then check whether the actual deal economics clear the bar before you commit.