Examiner Mindset Guide

#WBS11 — Examiner Mindset Guide (Unit 1: Marketing and People)

Provenance (2026-09-29). Built from a direct read of research/veridian/WBS11-verified-facts.md §2 (the verbatim level-descriptor tables and the Business-specific mark-scheme architecture) and signal-examiner-intelligence.md (read first, so nothing below repeats its 16-series read of examiner-report phrasing and format patterns), plus two mark schemes and their paired examiner's reports freshly fetched and re-extracted this session (pdftotext -layout) from the local archive at ~/Downloads/PAPERI/IAL_Unit1_Complete_2024_2025/IAL_Materials/Business/, whose legitimacy the facts bank's own §0 already established: January 2023 (mark scheme, Publications Code WBS11_01_2210_MS, Question Paper Log Number P72451A; examiner's report, Publications Code WBS11_01_ER_2301) and June 2022 (mark scheme, Publications Code WBS11_01_2206_MS, Log Number P69456A; examiner's report, Publications Code WBS11_01_2206_ER). The January 2023 mark scheme's own printed code genuinely reads 2210 (an internal Pearson batch-numbering quirk, not a transcription error here — independently confirmed against the question paper's own Log Number and its Realme-smartphone Q1(e), which matches the mark scheme's own question content exactly). Not affiliated with or endorsed by Pearson Edexcel. Every quotation below is a short, attributed fragment (under ~15 words); no mark-scheme table, full question or full indicative-content list is reproduced.

#What this document is, and isn't

This is not a second read of the examiner-report prose — signal-examiner-intelligence.md already did that across 16 series, and found the reports' own rhetorical patterns (stock paragraphs, the "generic" verdict word, the June-vs-January/October format split). This document looks instead at the mark scheme's own printed instructions to the examiner — the general marking guidance and the per-question "indicative content guidance" that governs how a script is actually marked, before any examiner ever writes a report about it — and at real annotated marking decisions from the archive, to show that machinery working on an actual script. It is also not a second pass at level-boundary language: forge-level-threshold-guide.md already diffs each ladder's descriptor wording level by level and pins it to real graded answers. That document's own §1 states, correctly, that the paper runs two mechanisms (discrete marks vs levels of response) — this document takes that distinction further, into how each mechanism is actually applied by an examiner reading a live script, using the mark scheme's own stated rules rather than inference from outcomes.

---

#1. Two marking mechanisms, confirmed from the mark scheme's own printed rules — not assumed

WBS11 does not mark every question the same way, and the mark scheme says so explicitly rather than leaving it to be inferred.

The 2-, 4- and 6-mark questions (Define, Calculate/Construct, Explain, Analyse) are marked by discrete point allocation against a fixed Knowledge/Application/Analysis split, printed at the top of each question's answer block. January 2023's Q1(b) ("Explain how the data above will affect the revenue of the retailer") is headed, verbatim, "Knowledge 1, Application 2, Analysis 1" [MS-Jan23] — four independently-awarded marks, each either present in the response or not. There is no band, no "best fit," no judgement about which of two adjacent categories an answer sits closer to: a candidate either wrote the second creditable application point or they didn't.

The 8-, 10- and 20-mark questions (Discuss, Assess, Evaluate) are marked holistically against a single level-descriptor ladder — and the mark scheme states, per question, that this is not a checklist exercise. Every level-marked question in both mark schemes checked this pass carries the identical instruction, word for word, immediately above its indicative content: "Answers must be credited by using the level descriptors... in line with the general marking guidance" [MS-Jan23, MS-Jun22]. The indicative content itself is then explicitly demoted from a scoring list to exemplification only: it "does not imply that any of these must be included" [MS-Jan23, MS-Jun22], and "other relevant points must also be credited" [MS-Jan23, MS-Jun22]. This exact three-sentence instruction recurs, unchanged, under Q1(d), Q1(e) and Q2(d)/Q2(e) within the June 2022 mark scheme alone — it is not a one-off caveat but the standing rule for every levels-based question on the paper. This confirms directly, from the primary document rather than from inference, what WBS11-verified-facts.md §7 already concludes from a different angle (Business's levels-based questions carry one indicative-content list marked against one level ladder, not a separate knowledge/application/analysis band plus a separate evaluation band the way some other levels-marked subjects split it): the "indicative content" a candidate sees echoed in a mark scheme is a menu of acceptable routes to a level, not a set of boxes to tick, and a response that never uses any of the printed bullets can still reach Level 4 on the strength of an equally valid point the mark scheme's own rule requires the examiner to credit anyway.

Practical consequence for a candidate: for Define/Calculate/Explain/Analyse, hitting the right number of distinct, creditable points is what matters — a technically correct answer that only makes one point where two are needed is capped by the split itself, not by examiner generosity. For Discuss/Assess/Evaluate, there is no equivalent "did I hit N points" arithmetic — the question the mark scheme obliges the examiner to ask is which level descriptor the answer as a whole best matches, and a genuinely original, well-argued point absent from the exemplar list is credited exactly as if it had been printed there.

---

#2. What examiners are told to do before they read a single word of a script

Every mark scheme carries a "General Marking Guidance" section before Q1(a) — administrative rules, not exam content, so they can be discussed more fully than the exam questions themselves. Confirmed identical, word for word, in both series checked this pass (June 2022 and January 2023), which is consistent with this being standard Pearson-wide boilerplate rather than a WBS11-specific policy — but it is nonetheless the literal instruction this paper's examiners mark under, and nothing in the examiner-report prose ever states these rules directly; the reports only report outcomes.

• Consistency across the whole entry, not just within one examiner's stack. "Examiners must mark the first candidate in exactly the same way as they mark the last" [MS-Jan23, MS-Jun22].
• Positive marking — reward presence, don't penalise absence. Candidates "must be rewarded for what they have shown they can do rather than penalised for omissions" [MS-Jan23, MS-Jun22]. This is the mechanical root of something signal-examiner-intelligence.md §3 already notices from the outside (that a January/October report's most common complaint is a missing half of the answer, not a wrong one): the marking instruction is built to score what's there, which is exactly why an unbalanced, one-sided response still scores something — it is being marked for what it contains, then capped by what it lacks (the level descriptors' own escalating counterbalance requirement), never marked down as if the omission were itself a fault.
• Mark to the scheme, not to a guessed grade boundary. Examiners are told explicitly not to mark "according to their perception of where the grade boundaries may lie" [MS-Jan23, MS-Jun22] — the two are institutionally firewalled from each other in the marking instruction itself.
• No artificial ceiling. "There is no ceiling on achievement" [MS-Jan23, MS-Jun22]; full marks are to be awarded whenever a response genuinely matches the scheme, with zero equally available when nothing in the response is creditable.
• Escalation, not guesswork, when a response is ambiguous. "Where some judgment is required... exemplification may be limited" [MS-Jan23, MS-Jun22], and in that case "the team leader must be consulted" [MS-Jan23, MS-Jun22] rather than the marking examiner deciding alone. This is the one rule with a direct, practical implication for how a script should be written: an answer that forces an examiner into genuine doubt about which level it fits is an answer that has made its own marking harder, not more generous.
• Crossed-out work is still marked unless replaced. Work a candidate crosses out is marked "unless the candidate has replaced it with an alternative response" [MS-Jan23, MS-Jun22] — so a candidate who crosses out a first attempt and leaves nothing in its place has not necessarily forfeited it.

---

#3. The mechanism in action: three real, annotated marking decisions from June 2022

June-series examiner's reports reproduce real (transcribed) candidate scripts with a boxed examiner comment stating the mark and the reason — the one format, per signal-examiner- intelligence.md §4, that lets you see the level ladder actually being applied rather than just described. Three from June 2022, read together, show the order in which the checks in §§1–2 above actually get used on a live script.

Check one: does this answer the question that was actually asked — checked before content quality, and capable of capping a response regardless of it. Q1(d) asked candidates to discuss the impact of competition on passengers. One response was "awarded Level 1 — 2 marks" despite the examiner's own comment that "the candidate shows some knowledge of the topic" [PEF-Jun22] — capped there specifically because the candidate wrote about "the impact of competition on passengers and not the impact on the business" [PEF-Jun22]. Accurate knowledge of competition, on its own, did not move the response off Level 1, because the level descriptors' own bottom rung already assumes "isolated elements of knowledge" can be present — what keeps a response at Level 1 despite real content is that the content answers a different, adjacent question.

Check two: once the target is right, which level's own descriptor language does the answer's actual content match, clause by clause? Q2(e) asked candidates to assess product orientation. The Level 4 exemplar's examiner comment tracks the descriptor almost clause for clause: "supported by relevant and effective use of business context" [PEF-Jun22], then "balanced with wide-ranging counter arguments and leads to a supported judgement" [PEF-Jun22] — language that is not the examiner's own summary so much as a direct check against the Level 4 descriptor's own listed requirements (accuracy and context, then balance, then a supported judgement), confirmed present in that order.

Check three: within the level, how completely are the descriptor's own qualifiers met — because a real attempt at the top requirement can still fall short of it. Q3's democratic-leadership essay produced a response "given high Level 3 marks" specifically because, although "counterbalance is attempted," the examiner judged "the points are a little repetitive and assertive in nature" [PEF-Jun22] — an attempt at the Level 4 requirement (a full awareness of competing arguments) that falls short of it not through absence but through under-development, which is exactly the distinction the level descriptors are built to separate and the discrete K/App/An split in §1 has no equivalent mechanism for capturing.

Read in sequence, these three decisions are the mark scheme's rules in §§1–2 applied to real scripts in the order they actually bind: target first (does this answer what was asked, at all), then match (which level's own language fits the content that's there), then completeness within that match (how fully the level's own top qualifier — usually balance, or a supported judgement — is actually met). This gives one further layer under signal-examiner-intelligence.md §10's own synthesis (that the examiner reads above all for whether an answer visibly answers the question asked): that check is not just the report-writer's most common complaint, it is the literal first gate the mark scheme's own level descriptors apply, because a response answering the wrong target cannot climb past the bottom rung no matter how accurate its content, exactly as Q1(d) shows.

---

#4. What this means practically

1. For Define/Calculate/Construct/Explain/Analyse, count creditable points, not paragraphs. The K/App/An split in §1 means length beyond the number of genuinely distinct, creditable points does nothing — a fourth sentence restating the third point scores the same as a third sentence that stops there.
2. For Discuss/Assess/Evaluate, answer the actual target of the question before developing the argument. §3's Q1(d) example shows accurate content capped at the bottom of the ladder when it answers an adjacent question rather than the one asked — this happens before the examiner ever reaches the balance/counterbalance check that the rest of this hub's error tracker already documents as the paper's most common mid-to-high-level trap.
3. A genuinely relevant point not on the mark scheme's own printed list is still credited. §1's "indicative content guidance" is a real, per-question, mark-scheme-printed instruction to the examiner, not a hopeful assumption — a candidate is not disadvantaged for reasoning to a valid point the exemplar list happens not to include.
4. An answer should resolve its own ambiguity, because the mark scheme's fallback for ambiguity is escalation, not leniency. §2's team-leader-consultation rule exists for cases the marking examiner cannot confidently place — which is a reason to write clearly enough to be placed, not a safety net a candidate can rely on.

---

#5. Gaps flagged, not filled

• How examiners are actually trained and standardised (standardisation meetings, moderation procedures, how team leaders resolve an escalated case) is not documented in anything reviewed for this paper — Pearson does not publish that material, and nothing above claims insight into it beyond the general marking guidance that is printed in the mark scheme itself. This document is confined to what that printed guidance and the annotated exemplars actually show.
• Cross-series stability of the general marking guidance was checked in two series only (June 2022, January 2023), both confirmed identical. That is enough to state the wording is not paper-specific to one series, not enough to claim it has never changed across the paper's whole history (January 2019 – present) — no claim of that breadth is made above.
• No grade boundary, likelihood percentage, or "confirmed in every series" claim appears anywhere in this document, consistent with this hub's standing rule: none of the source material reviewed for it contains one.

---

#Sources cited

Both series fetched from the local archive at ~/Downloads/PAPERI/IAL_Unit1_Complete_2024_2025/IAL_Materials/Business/, re-extracted this session with pdftotext -layout, and cross-checked internally (question paper log number against mark-scheme log number, and question content — the Realme smartphone stimulus — against the question paper itself) rather than trusted from filename alone.