Topic Brief: Decision Trees Cpa and Contribution

*Provenance (checked 2026-09-28): research/veridian/WBS13-verified-facts.md read in full and grepped for every decision-tree/CPA/contribution hit (spec 3.3.3.3–3.3.3.5), including the full §3.3.3 write-up and its provenance table; content/veridian/curriculum/wbs13/decision-trees-cpa-and-contribution.ts read in full (its worked chains — Center Parcs, Thornfield Furniture Co., Solmere Outdoors, Aldergate Print Co. — are cross-referenced, not reproduced); content/veridian/technique-hub/wbs13/codex-deep-dive-decision-trees-cpa- and-contribution.md read in full, including its own Part 4 cross-topic table — that table is not repeated here verbatim; this brief adds two syntheses the CODEX document does not make (the extract-lettering pattern and the Section B/C placement split, both in §3 below); signal-frequency-and-boundaries.md read in full for its 12-series denominator convention and paper-structure facts; forge-exam-technique-guide.md read for the Section A/B/C definitions used below; content/veridian/question-bank/wbs13.ts grepped for this topic's terms — zero hits, nothing to duplicate. No new primary-source PDF was fetched for this pass; every fact below traces to the two research documents named first, cited inline by series and question.

Reading-depth caveat: twelve WBS13 series are located for this course (Oct 2021–Oct 2025), but the verified-facts file's own targeted decision-tree/CPA/contribution keyword search — the pass that actually found the hits below — was run against nine of them (Oct 2021, June 2022, Oct 2022, Jan 2023, June 2023, Oct 2023, Summer 2024, Oct 2024, Jan 2025); the remaining three (Jan 2022, Jan 2024, Oct 2025) were mined for other sub-topics but a dedicated keyword pass for this one is not confirmed in the research file. The counts below still use 12 as the denominator, matching the standing convention every other WBS13 SIGNAL document in this folder uses — treat each count as a floor, not a ceiling, per that convention's own stated caveat.*

#WBS13 SIGNAL — Topic Master Brief: Decision Trees, CPA and Contribution

Spec 3.3.3.3–3.3.3.5. Three separate spec points sharing one lesson file and one CODEX deep-dive. This brief does neither job again — mechanism and worked numbers live in decision-trees-cpa-and-contribution.ts, theory and real-exam write-ups live in the CODEX deep-dive. What follows is the pattern across all three real hits taken together, framed as revision guidance rather than restated as reference material.

#1. Every real series confirmed testing this topic

Seen in 4 of 12 series checked, one confirmed question each for decision trees and contribution, two for critical path analysis (CPA is the only one of the three confirmed more than once):

Real figures behind each: Coca-Cola's project duration = 32 weeks; Lush's total contribution on 3,300,000 units = £14,850,000; Burger King's project duration = 11 weeks with Activity C carrying 1 week of float; Center Parcs' decision tree = €80m net gain (new holiday village) versus €33m (takeover), both independently recomputed and confirmed against the mark scheme's own working.

#2. The mark-tariff pattern

Every confirmed decision-tree or CPA question is the 20-mark Evaluate essay — Section B (Q2) or Section C (Q3) — never a Discuss (8) or Assess (12). Contribution sits at the opposite end of the tariff scale: the one confirmed hit is a 4-mark Calculate opening Section A. No confirmed mid-tariff (6- or 8-mark) question exists for any of the three sub-topics in the series checked.

Section placement is not fixed to one slot. Of the three 20-mark hits, two landed in Section B (Coca-Cola CPA, Center Parcs decision tree) and one in Section C (Burger King CPA) — worth knowing precisely because it rules out an assumption that this topic group always sits in one section rather than the other.

#3. Recurring context/angle

Two patterns hold across all four real hits that neither the lesson file nor the CODEX deep-dive states as a named pattern:

Every confirmed decision-tree or CPA question is built from a pair of adjacent, late-alphabet extracts — Coca-Cola from Extracts F–G, Center Parcs from Extracts E–F, Burger King from Extracts G–H — never from the earlier A–D extracts that Section A's calculations draw on. If a real paper's Section B or C essay names two extracts from late in the source booklet, that alone is a stronger signal this topic group is in play than the topic name appearing in the question stem.

The calculation is the evidentiary base the evaluation is built on, not a separate hurdle before it. All three 20-mark hits require a real number worked from extract data — expected monetary value for the decision tree, project duration and float for both CPA questions — and both CPA examiner reports confirm the higher mark bands went to answers that evaluated the technique as applied to that specific named business (Burger King's multi-site rollout; Coca-Cola's own relaunch), not as a generic list of textbook benefits and limitations.

All four real businesses are large, named, real-world brands — Coca-Cola, Lush, Burger King, Center Parcs — consistent with the wider pattern already recorded across 3.3.3 as a whole (Pets at Home, Spotify, Airbus/IAG and Brompton Bikes also appear on this paper's other quantitative sub-topics).

#4. If this topic appears — what to expect

Most likely shape, based on what has actually happened in 4 of 12 series: if decision trees or CPA is the technique behind a 20-mark Section B or C essay, expect two adjacent extracts from late in the source booklet carrying the option or activity data, a required numeric calculation (EMV, or project duration plus at least one activity's float) that the mark scheme treats as the evidence for the judgement rather than an optional add-on, and a top band reserved for evaluation grounded in the named business's own situation. If contribution appears instead, the confirmed precedent is far more modest: a short Section A Calculate question built from a straightforward selling-price/variable-cost pair, with two independently valid working methods both creditable.

Two things stay genuinely unconfirmed and should not be assumed: no primary-source WBS13 mark scheme has supplied a machine-readable EST/LFT/float activity table for either CPA hit — both underlying networks exist only as diagram images — so a worked network at the exam's own numbers cannot be reproduced, only the totals and named activities already listed above; and no real WBS13 question has been found testing contribution specifically as a decision-making technique (accept/reject, discontinuation, or scarce-resource ranking, 3.3.3.5c) — only the Lush calculation itself is real-exam-sourced. Nothing here is a likelihood or grade-boundary claim; "seen in N of 12 series checked" is the complete, honest statement this document makes.