9.4 Agile Budgeting, Contracts & Governance

Agile Budgeting, Contracts & Governance

The Failure of Traditional Annual Budgeting

Traditional enterprise budgeting operates on a rigid 12-to-18-month cycle. In the autumn of each year, business units write speculative 50-page business cases forecasting costs, revenues, and resource requirements for projects that will not even begin for another six months. Finance departments review these business cases, lock in fixed budgets, and allocate funds to temporary projects.

This predictive budgeting model collapses under the conditions of complex knowledge work, generating systemic dysfunctions:

1. The "Spend-It-or-Lose-It" Pathology: Departmental managers know that if they do not spend 100% of their allocated annual budget, finance will permanently reduce their baseline allocation in the next fiscal year. Consequently, in the fourth quarter, managers rush to authorize low-value, wasteful expenditures simply to exhaust remaining funds.
2. Funding Temporary Projects Instead of Stable Teams: Traditional accounting funds "projects" with defined start and end dates. When a project begins, a temporary team is assembled from various departments. When the project ends, the team is disbanded, scattering institutional knowledge, breaking psychological safety, and resetting the team development curve to zero.
3. Lock-In to Obsolete Hypotheses: Market realities, competitor moves, and customer preferences change rapidly. An annual budget locks capital into initiatives conceived up to 18 months earlier, making it politically and administratively difficult to pivot capital toward emerging opportunities.
4. Disincentivizing Early Completion: If an agile team delivers 90% of the desired business value in 3 months instead of the planned 12 months, traditional financial systems penalize them by canceling the remaining project budget and reallocating resources, rather than rewarding early value realization.

Dynamic and Incremental Funding: Beyond Budgeting in Practice

To overcome these failures, agile enterprises adopt principles from the Beyond Budgeting movement and modern venture capital:

1. Capacity Funding for Persistent Teams

Instead of funding temporary projects, agile organizations fund dedicated, persistent, cross-functional teams (or value streams). The cost of a stable Scrum Team is highly predictable:

$\text{Team Run Rate} = \text{Headcount} \times \text{Blended Monthly Labor Cost} + \text{Fixed Tooling/Infrastructure Cost}$

Because team capacity and cost are stable, the enterprise stops estimating project costs through speculative labor hour calculations. Instead, the organization prioritizes a single, ordered backlog of strategic initiatives and brings the highest-value work to the persistent team.

2. Venture Capital-Style (Tranche-Based) Incremental Funding

Rather than granting a multi-million-dollar annual budget upfront based on a speculative business case, leadership acts like an internal venture capital firm:

• Seed Funding: An initiative receives a small initial tranche of funding (e.g., funding 3 Sprints) to build a prototype, test customer desirability, and validate core architectural assumptions.
• Empirical Governance Reviews: At the conclusion of each funding tranche, leadership, the Product Owner, and stakeholders inspect the working Increment and review Evidence-Based Management (EBM) metrics:
  • Current Value (CV): Are active users adopting the feature? Is customer satisfaction improving?
  • Unrealized Value (UV): What remaining market potential justifies further investment?
  • Time-to-Market (T2M): How rapidly are new releases reaching production?
  • Ability to Innovate (A2I): What proportion of budget is spent on new capabilities versus defect maintenance?
• Pivot, Persevere, or Stop: If the initiative demonstrates strong value metrics, the next funding tranche is authorized. If the hypothesis fails, leadership terminates the initiative early with minimal financial loss, redirecting surplus capital to high-potential opportunities.

Agile Contracting Models: Aligning Incentives Under Uncertainty

Procurement departments often attempt to force agile software development into traditional Fixed-Price / Fixed-Scope contracts. This creates an adversarial relationship between buyer and vendor:

• The buyer writes rigid, exhaustive specifications to lock down scope.
• The vendor inflates estimates by 50% to 100% to buffer against unforeseen risks.
• During development, when requirements inevitably change, the relationship devolves into bitter "Change Order Wars." The vendor demands exorbitant fees for minor scope modifications, and quality is compromised to protect profit margins.

Agile contracts replace adversarial tension with shared risk, transparency, and empirical alignment:

Deep Dive: "Money for Nothing and Changes for Free"

Formulated by Dr. Jeff Sutherland (co-creator of Scrum), this contracting model provides an ideal balance of flexibility and commercial safety:

1. Changes for Free:
   • At the beginning of any Sprint Planning session, the client can introduce new Product Backlog items or modify existing ones at zero additional charge.
   • The Ground Rule: For every new item added, the client must remove an unstarted item of equivalent estimated effort. The overall scope volume remains constant, accommodating emergent learning without contract amendments.
2. Money for Nothing:
   • In software, the final 20% of planned features often represent low-value edge cases that deliver negligible return on investment.
   • If the client determines that the core business value has been fully achieved early (e.g., after Sprint 6 of a planned 10-Sprint contract), the client has the contractual right to terminate the contract early.
   • The Financial Settlement: The client pays a negotiated early-termination fee—typically 20% of the remaining unbilled contract balance—and walks away with 100% of the working, tested software.
   • The Win-Win: The client saves 80% of remaining expenditures ($80,000 saved on a remaining $100,000 balance). The vendor immediately banks $20,000 in pure profit margin without spending any labor or overhead, freeing their engineering staff to take on new billing client engagements.

Governance Without Stage Gates

Funding and contracts are two of the three enterprise interfaces where traditional structures collide with Scrum; the third is governance. Empirical governance replaces document-based stage gates with frequent inspection of Done Increments:

• Investment decisions at the Sprint Review cadence. Leaders inspect working software and evidence (EBM measures) at the end of each funding tranche and decide whether to persevere, pivot, or stop.
• Compliance inside the Definition of Done. Regulatory, security, and audit requirements become part of the Definition of Done and are automated in the delivery pipeline, so every Increment carries its own evidence. Section 7.3 covers this in depth for regulated environments.
• Evolve, don't abolish. Most organizations cannot drop annual budgeting or procurement rules in one step. The Scrum Master makes the cost of each traditional control transparent and helps leaders replace it incrementally—for example, keeping the annual budget but releasing it in quarterly tranches tied to inspected results.