6.1 When to Use a Predictive Approach
Key Takeaways
- Predictive (waterfall) approaches fit best when requirements are well-defined, stable, and unlikely to change, and the technology is proven
- Domain 2 (Predictive, Plan-Based Methodologies) is 17% of the 150-question CAPM exam, so roughly 23-26 scored items test this material
- High regulatory, compliance, and fixed-price-contract environments favor predictive because scope must be defined and documented up front
- The Stacey Complexity Model maps requirements certainty against technology certainty to recommend predictive, hybrid, or adaptive approaches
- Predictive trades flexibility for predictability: clear milestones and easy cost/schedule estimation, but late feedback and inflexibility to change
Where This Fits on the Exam
The CAPM exam, updated in July 2023, is built on a four-domain Exam Content Outline (ECO): Project Management Fundamentals (36%), Predictive, Plan-Based Methodologies (17%), Agile Frameworks/Methodologies (20%), and Business Analysis Frameworks (27%). You sit 150 questions (135 scored plus 15 unscored pretest items) in 3 hours, split into two 75-question sections by a scheduled ~10-minute break (you cannot revisit the first 75 once you pass it).
Because Domain 2 is 17% of scored content, expect roughly 23 questions drawn from this chapter — so judgment about when predictive fits, plus the schedule mechanics later in the chapter, are high-value.
The predictive approach — also called waterfall or traditional — moves through largely sequential phases, each substantially complete before the next begins, with formal phase-gate reviews between them:
Requirements -> Design -> Build -> Test -> Deploy -> Close
When Predictive Works Best
Predictive is the right call when you can define what you are building before you start building it. The exam frames this through stable requirements, low uncertainty, and a need for documentation.
| Factor | Predictive is appropriate when... |
|---|---|
| Requirements | Well-defined, documented, and stable from the start |
| Scope | Fixed and unlikely to change significantly |
| Technology | Proven, well-understood, low technical risk |
| Uncertainty | Low — few unknowns about what or how |
| Regulations | High compliance demands extensive documentation and traceability |
| Contracts | Fixed-price, where scope must be defined and priced up front |
| Stakeholders | Prefer defined milestones, deliverables, and a single final handover |
When predictive is the wrong fit
- Requirements are unclear, evolving, or expected to change frequently
- The product has never been built before (high innovation/discovery)
- Technology is new or unproven, creating high technical uncertainty
- Early and continuous customer feedback is essential to success
- Time-to-market is critical and incremental delivery adds value sooner
Common trap: Candidates pick "predictive" simply because a project is large or expensive. Size is not the deciding factor — stability of requirements and certainty of technology are. A large, well-understood construction project is predictive; a large, exploratory R&D project is not.
The Stacey Complexity Model
The Stacey Complexity Model (Stacey Matrix) helps choose an approach by plotting how clear the requirements are against how certain the technology/how is.
| Requirements CLEAR | Requirements UNCLEAR | |
|---|---|---|
| Technology CERTAIN | Simple -> Predictive | Complicated -> Hybrid |
| Technology UNCERTAIN | Complicated -> Hybrid | Complex/Chaotic -> Adaptive (Agile) |
- Simple (clear + certain): use predictive
- Complicated (uncertainty in one dimension): often hybrid
- Complex (uncertainty in both): use adaptive, embracing change through iteration
- Chaotic (extreme uncertainty): adaptive with rapid experimentation
Organizational Structure Fit
Predictive aligns naturally with hierarchical, plan-driven cultures.
| Structure | Fit | Why |
|---|---|---|
| Functional | High | Hierarchical control, specialized resource pools |
| Weak matrix | Moderate-High | Functional managers keep authority |
| Balanced/Strong matrix | Moderate | Shared or PM authority works in structured work |
| Projectized | Moderate | Full PM control, but less needed when scope is stable |
Advantages vs. Limitations
| Advantages | Limitations |
|---|---|
| Clear structure and defined milestones | Inflexible once baselines are set |
| Cost and schedule are easier to estimate | Issues surface late (testing is near the end) |
| Comprehensive documentation for compliance | Customer sees the product only at delivery |
| Works well with fixed-price contracts | Assumes requirements are complete and correct |
| Straightforward progress measurement | Long cycles before value is realized |
Worked scenario: A bank must rebuild a regulatory reporting system to a published government specification by a hard statutory deadline, under a fixed-price vendor contract. Requirements are dictated by regulation (clear) and the reporting technology is mature (certain). On the Stacey Matrix this is simple, so the exam-correct answer is a predictive approach with strong documentation and change control — not agile experimentation.
Development Life Cycles to Recognize
The CAPM distinguishes several life cycles, and Domain 2 questions often ask you to name the predictive end of the spectrum versus its neighbors.
| Life cycle | Scope | Delivery | Change |
|---|---|---|---|
| Predictive | Fixed early | Once, at the end | Restricted, via change control |
| Iterative | Dynamic | Refines in repeated cycles | Expected between iterations |
| Incremental | Dynamic | Delivers usable pieces over time | Expected each increment |
| Adaptive (agile) | Dynamic | Frequent small increments | Embraced continuously |
| Hybrid | Mixed | Combines predictive + adaptive | Phase-dependent |
A predictive life cycle determines scope, time, and cost in the early phases, which is exactly why it depends on requirements being knowable up front. If the exam describes a project that delivers its full product in a single release after sequential phases, that is the predictive (waterfall) life cycle. If it describes building and shipping usable slices over time, that is incremental — a frequent distractor.
Which project is MOST suited for a predictive (waterfall) approach?
According to the Stacey Complexity Model, which approach is recommended when BOTH requirements and technology have high uncertainty?
A key limitation of the predictive approach is that: