Chains Guide

Provenance (written 2026-09-29): research/veridian/WBS13-verified-facts.md (180KB, read in full) was checked first for every "chain" reference. Every mark-scheme and examiner-report quotation below was then re-extracted directly from the primary-source PDFs rather than trusted from that file's own transcription — Business_Unit 3_MS/QP/ER_January_2023.pdf and Business_Unit 3_MS/QP/ER_October_2023.pdf, both held locally at /Users/neosteinhoff/Documents/SOL/Papers/IAL_Paper3_Complete/IAL_Materials/Business/, re-run through pdftotext -layout for this pass. No FINSTO source lead exists for a WBS13 chains guide (only WBS12 and WEC12 versions were found in ~/Downloads/FINSTO/), so nothing here is a verification-against-a-flawed-draft exercise — it is built directly from the primary documents. content/veridian/curriculum/wbs13/influences-on-business-decisions.ts and content/veridian/technique-hub/wbs13/codex-complete-worked-paper.md were both read in full first so this file stays additive: the Stellantis CEO-pay question already has extensive teaching prose, a stakeholder comparison table and two flashcards in the curriculum lesson, and the Five Guys financial-rewards question already has a complete graded Level 1–4 worked answer in the CODEX paper — neither is reproduced here. This document's own job, not done anywhere else in this course's WBS13 material, is a dedicated treatment of the chain mechanism itself: what Pearson's own mark-scheme language actually requires sentence-by-sentence, verified against this paper's own real tariff structure, which differs from WBS12's in ways worth stating plainly rather than assumed. Quotations from Pearson material are kept short (under fifteen words each) and cited by series and question so they can be checked independently.

#WBS13 Chains Guide — How a Point-Development Chain Actually Earns Marks on This Paper

Every levels-based question on this paper — every Discuss, every Assess, every Evaluate — is marked against a descriptor that uses the word "chain" or "chains" to separate the bands. That is Pearson's own mark-scheme vocabulary, not a tutor's shorthand, and it is checked directly below against two real papers rather than assumed to work the same way it does on WBS12. It doesn't, in one important respect: this paper's own tariff structure is different, and that difference changes what a chain actually has to do to earn marks here.

---

#1. Why "chain" is not a teaching metaphor — and why WBS13's tariffs aren't WBS12's

Checked directly against the October 2023 mark scheme (Business_Unit 3_MS_October_2023.pdf, re-extracted for this pass):

• 8-mark Discuss, Level 2: "Chains of reasoning are presented, but may be assertions or incomplete."
• 8-mark Discuss, Level 3: "Analytical perspectives are presented, with developed chains of reasoning" — showing cause(s) and/or effect(s).
• 12-mark Assess, Level 4: "A coherent and logical chain of reasoning, showing cause(s) and/or effect(s)."

And from the January 2023 mark scheme (Business_Unit 3_MS_January_2023.pdf, same direct check), the 20-mark Evaluate's top band:

• Level 4: "well-developed and logical, coherent chains of reasoning" — showing a range of cause and/or effect(s).

This is the real, current wording, not a paraphrase. But three structural facts change what that wording means in practice on this paper, and none of them can be assumed from WBS12:

First: Assess is a 12-mark tariff here, not 10. Confirmed directly on two real questions in the same paper — Oct 2023 Q1(d) ("Assess the likely effectiveness for Five Guys of using financial rewards to reduce labour turnover") and Q1(e) (Ansoff's Matrix) — both run Level 1 (1–2), Level 2 (3–4), Level 3 (5–8), Level 4 (9–12). The spec's own Appendix 6 table lists two different Assess tariffs — 10 marks for Units 1/2, 12 for Units 3/4 — and the higher figure is the one this paper actually uses.

Second: the spec lists a 6-mark Analyse tariff ("a brief chain of reasoning… does not include evaluation") that this course's research has never found actually sat. Eleven series have been located and at least partially mined for this paper (Oct 2021, Jan 2022, June 2022, Oct 2022, Jan 2023, June 2023, Oct 2023, Jan 2024, Summer 2024, Jan 2025, Oct 2025); every one that has had its Section A tariff structure checked runs four short Calculate/Construct questions and two Discuss/Assess questions summing to 40 marks — the pattern this pass independently confirmed on Oct 2023 is 4+4+8+12+12. No 6-mark Analyse question has turned up in any of them. That is an absence across every series checked, not a claim that no such question has ever existed on any WBS13 paper — but it means the version of this document that would matter for revision is: this paper's real chains are built for Discuss (8), Assess (12) and Evaluate (20) only, and any "AO1×2/AO2×2/AO3×2" Analyse breakdown (the kind that genuinely exists and matters on WBS12) is not confirmed to be relevant here at all.

Third: marking is holistic on every one of those three real tariffs, checked in the same document's own general guidance. From the Oct 2023 examiner's report, verbatim: mark schemes are "applied in a holistic way rather than looking for individual Assessment Objectives." There is no per-AO mark tally to hit on a Discuss, Assess or Evaluate question the way there is on Q1(a)/(b)'s 4-mark Calculate questions (which the same mark scheme labels plainly: "Knowledge 1, Application 2, Analysis 1" — a real, discrete three-part structure, but a calculation mark scheme's structure, not the mark-scheme's own "chain of reasoning" language, and the two should not be conflated).

---

#2. What a chain is built from on this paper

Stripped to what these two real mark schemes actually reward, a chain on WBS13 has the same four real components as any other paper that uses this descriptor language — the names are this course's own teaching labels, not Pearson's, but each is anchored to a quoted phrase:

Two further requirements, both directly checked rather than assumed: a conclusion is explicitly not wanted on Discuss ("A conclusion is not required for 8-mark questions" — Oct 2023 ER, Q1c commentary) but is explicitly wanted on both Assess and Evaluate ("A conclusion/judgement is required for 12 mark questions but was not often seen" — Oct 2023 ER, Q1d commentary). A chain that stops at Mechanism → Outcome is a complete Discuss chain but an incomplete Assess or Evaluate one.

---

#3. Two real chains, two real papers

#Chain 1 — Stellantis, the CEO pay-rise vote (Jan 2023 Q2, 20-mark Evaluate)

The real question (Business_Unit 3_QP_January_2023.pdf, extracted directly), using Extracts E and F, asks candidates to evaluate "whether the board of directors of Stellantis should accept the shareholders' vote."

Verified extract data, read directly from the question paper rather than the mark scheme's own paraphrase of it: at an April 2022 investors' meeting, "just over 52%" of Stellantis shareholders voted against a proposed 17.6% pay rise for CEO Carlos Tavares, taking his pay to €19m. Volkswagen's CEO was due €8.6m for the same year; Ford's Jim Farley was due "nearly $23m (€22m)." Stellantis's 2021 revenue was €149,419m against €47,656m in 2020; profit for the year was €14,208m against €2,023m; earnings per share rose from €1.41 to €4.64; employee numbers fell from 298,879 to 281,595. One detail this pass found in the extract itself that is not in this course's existing Stellantis material anywhere — not the curriculum lesson, not the research file: "The company's share price remained unchanged at €13.97" after the vote.

A worked chain built from those verified facts (this course's own construction, not a Pearson quotation):

One reason the Board might accept the shareholders' vote and reject the pay rise is that the financial case for restraint is real, not just symbolic — Volkswagen's CEO was paid only €8.6m for the same year, and Stellantis's own workforce fell by over 17,000 people since 2020 without a comparable rise of its own, so awarding Tavares a 17.6% increase to €19m risks a genuine demotivating effect on employees who can see the gap. However, this has to be weighed against Stellantis's 2021 results, which were exceptional under Tavares specifically — revenue up 213.54%, profit up 602.32%, earnings per share more than tripling from €1.41 to €4.64 — and against a directly comparable peer, Ford's Jim Farley, on nearly €22m; against that comparator, €19m is not an outlier. This matters because the risk of rejecting the rise is not costless either: a CEO who feels undervalued can leave for a competitor, and Stellantis would then have to find a replacement capable of repeating that performance. On balance, the Board should likely accept the vote's underlying concern about pay restraint without treating it as a mandate to block the rise outright — especially since the vote itself was a narrow 52% against, meaning 48% of shareholders did not object, and the market's own verdict was silence: the share price sat unchanged at €13.97, suggesting investors did not read the vote as a crisis requiring the Board to act against its own CEO.

What the real examiner report says about this exact question (checked directly against Business_Unit 3_ER_January_2023.pdf): there was "no 'correct' answer" here — the report frames it as "about the quality of arguments from the shareholders viewpoint" against the CEO's. On what separated the bands: "candidates had to have developed chains and use of business theory" rather than a "'common sense' evaluation." The report also states plainly that "Extract F gave candidates plenty of financial reasons for the pay rise," and that weaker responses "paraphrased the information in the extracts" instead of building on it — the same underlying failure named below in §5, on a genuinely different question.

#Chain 2 — Five Guys, financial rewards and labour turnover (Oct 2023 Q1d, 12-mark Assess): what breaks this chain, not what a full answer to it looks like

This exact question already has a complete Level 1–4 worked answer in codex-complete-worked-paper.md and extensive real-mark-scheme teaching prose in assessing-competitiveness.ts — neither is repeated here. What belongs in a chains guide specifically, and is not built out fully anywhere else, is the mark-scheme's own account of the one mechanism error that breaks this exact chain, because it is a genuinely different failure mode from Chain 1's "paraphrasing the extract."

The real question, verbatim (checked directly against Business_Unit 3_MS_October_2023.pdf): "Assess the likely effectiveness for Five Guys of using financial rewards to reduce labour turnover."

The mark scheme's own indicative content gives the reason (Five Guys pays its top 200 restaurants "$900 to $1,300" a week when mystery-shopped) and the mechanism the question actually asks for: a regular, re-earnable reward gives staff a recurring reason to stay. But the real examiner's report records the specific way many candidates broke this exact chain — not by leaving data unused, but by building the wrong mechanism from the right reason: even some good answers still "focused on mainly productivity or motivation rather than labour turnover." A candidate who correctly explains that money motivates better performance has built a real, coherent chain — just not the one this question pays for. The reason (financial reward) was right; the Outcome named at the end of the chain (better service, not lower turnover) was the wrong outcome for this question's own command. That is a distinct, checkable trap from Chain 1's failure, worth naming as its own category in §5.

---

#4. Chain requirements by tariff — verified against this paper's own real mark schemes

The 4-mark Calculate tariff (Q1a/b on every series checked) carries its own discrete "Knowledge 1, Application 2, Analysis 1" structure, verified directly on Five Guys's ROCE calculation and NPV calculation in the same Oct 2023 mark scheme — a real three-part mark structure, but not one the mark scheme itself calls a "chain," and it is kept separate here rather than folded into the same label the levels-based questions use.

---

#5. Two real, distinct ways this paper's own examiners record a chain breaking

Copying the extract instead of building on it. From the Oct 2023 examiner's report's general guidance, not tied to one question: "Far too many candidates are still simply copying out large sections of the Extracts" with only "an attempt at limited evaluation," which "will only achieve lower levels." The Jan 2023 report names the same failure on Q2 specifically: weaker responses "paraphrased the information in the extracts" rather than building an argument from it. This is a failure to add anything past what the source already says — the data sits in the answer unconnected to a mechanism.

Building the right reason into the wrong mechanism. Distinct from the failure above: Chain 2's Five Guys trap is not that the mystery-shopper reward data went unused — it is that a real, coherent chain was built from it (financial reward → increased motivation → better performance), correctly reasoned, and simply aimed at the wrong outcome for this question's actual command word. The Oct 2023 examiner's report is explicit that this drew real credit for a real chain, just not enough of it: "To access the higher levels, the focus had to be on labour turnover," not on the productivity story the same facts could just as validly support.

Both are checkable against the documents cited above rather than against an invented occurrence count. No specific per-question error-frequency figure is stated in this document for either pattern, because none was found in the primary sources this pass checked directly.

---

#6. What this document did not attempt

This pass located and directly verified two real worked chains, one per tariff family (Evaluate, Assess), against real question papers, mark schemes and examiner reports read fresh for this document rather than trusted from the research file's own summary. It did not build a third worked chain for Discuss specifically, because the strongest candidate found (Oct 2023 Q1c, "Discuss the impact of this on its human resources") would have required re-deriving indicative content already adequately covered by this paper's own general Discuss requirements in §1 and §4 without adding a genuinely new mechanism — and this course's standing rule is that a document ends where the real, non-duplicative material runs out, rather than being padded to a target length with a third example that would not teach anything the first two don't already show.