SCOR DS kept the Level-2 layer and threw out the entire catalog — D4 has no successor code

The question

Verbatim, from the Notion Research Backlog:

"Did SCOR Digital Standard preserve the Level 2 process-type catalog (S1-S3 / M1-M3 / D1-D4), and where did D4 Deliver Retail Product go when Deliver split into Order and Fulfill?"

Context: this is a source-verification derivative of [[2026-08-17-gtm-motion-origins-industry-reference-models]]. That brief's structural argument — decomposition belongs at Level 1, the substitutable choice at Level 2, which is the model for the proposed axis + variants schema amendment — rests on exactly one primary document: the Supply-Chain Council's SCOR 6.0 Overview Booklet, circa 2003. The brief's own verify pass flagged its "stable through SCOR 12.0" claim as unsourced. If the current standard dropped or re-cut Level 2, the one solid leg gets shaky.

Answer, in one line: the Level-2 layer survives, every single Level-2 code does not. S1-S3, M1-M3 and D1-D4 are all gone — the letters S, M and D were retired or reassigned to different meanings — and D4 Deliver Retail Product has no successor category at any level. Retail-specific delivery was demoted to Level 4, which ASCM explicitly declines to standardize. Verified against two ungated ASCM primary PDFs for SCOR Version 14.0 (© 2025).

What we already know (from the vault)

• [[2026-08-17-gtm-motion-origins-industry-reference-models]] established, from a direct read of the SCOR 6.0 booklet, that Level 2 is the "Configuration Level (Process Categories)," that "a company's supply chain can be 'configured-to-order' at Level 2," and that the Execution row of the configuration toolkit reads S1-S3 / M1-M3 / D1-D4 including D4 Deliver Retail Product. That read is sound and is not disputed here — for the version it describes.
• The same brief flagged the gap this brief closes: "Whether the Level 2 process types survived into SCOR DS, and under what codes, is unresolved — no source found addresses it, and the primary documents were inaccessible." It also could not verify ASCM's open-access claim, recording "treat 'free' as reported, not verified."
• The brief's verify-vault-write gate listed as outstanding item 4: "'Stable through SCOR 12.0' (§2.2) is unsourced. The document actually read is SCOR 6.0. That claim converts a ~2003 booklet into a current ASCM standard and is doing heavy commercial work." That is the specific claim under test.
• [[2026-08-21-apqc-pcf-80-channel-motion-leaves]] is the sibling verification of the same research round, and it found the same gate-vs-asset asymmetry at APQC: the listing page is lead-gated, the static asset path is not. The identical trick worked at ASCM here.
• The live engagement stressing this taxonomy is Kwik Trip — convenience store, fuel, grocery, foodservice, plus vertically integrated manufacturing and distribution ([[2026-08-04-kwik-trip-ai-architecture-engagement-context]]). D4 was the closest thing in any reference model to an anchor for it, which is why this question was promoted.

What the web says

The route that worked, and the open-access claim is now verified

Two ASCM-hosted PDFs serve unauthenticated from ascm.org/globalassets/…, no login, no cookie, no referrer:

• scor-ds-digital-guide_final.pdf — the SCOR Digital Standard Quick Reference Guide, the full fold-out Level-0 → Level-3 process chart. 599 KB. Downloaded and text-extracted locally.
• intro-and-front-matter-scor-digital-standard-2025.pdf — the front matter, footered "CC BY-NC-ND 4.0. © 2025 ASCM — SCOR Version 14.0" on every page. 634 KB. Downloaded and text-extracted locally.

The front matter's own change log states: "Open access with authenticated ASCM login. There is no cost to register on ascm.org." The two assets above go further and need no login at all, and the model is published under a Creative Commons licence. The prior brief's "free-as-reported, not verified" caveat is closed: SCOR DS is genuinely open access, CC BY-NC-ND 4.0, no fee. The interactive full model at scor.ascm.org remains behind the free-account gate and was not opened.

apics.org (the SCOR 12.0 framework introduction PDF, the one document that would date the deletion precisely) returned HTTP 403 to both WebFetch and a browser-UA curl. Blocked, not exhausted-and-negative.

The Level-1 structure — confirmed, with a correction to how we've been stating it

Verbatim from the front matter: "The macro-level SCOR processes, include Level-0 and Level-1 processes. Orchestrate is a Level-0 process; Plan, Order, Source, Transform, Fulfill and Return are Level-1 processes and are well-known and widely adopted."

So the ordering we have been using — "Plan, Source, Transform, Order, Fulfill, Return" — is right on membership and wrong on sequence; ASCM's own order is Plan, Order, Source, Transform, Fulfill, Return, which is deliberate (order intake now precedes sourcing). Orchestrate sits at Level 0 and carries thirteen "Orchestration Enablers," OE1-OE13, covering supply chain strategy, business rules, performance, data/technology, human resources, contracts, network design, regulatory compliance, risk, ESG, enterprise business planning, segmentation, and circular supply chain.

The Level-2 catalog exists — here it is in full, from the primary chart

Every one of those codes is read directly off the ASCM Quick Reference Guide chart, along with their Level-3 children (O1.1-O1.6, F1.1-…, S1.1-S1.10, and so on).

The load-bearing observation: the letters were reused, not retired. S1 no longer means Source Stocked Product; it means Strategic Source. S2 is no longer Source Make-to-Order Product; it is Direct Procure. Anyone citing "SCOR S2" without a version is now ambiguous between two unrelated processes. M1-M3 are gone outright — Make became Transform and re-cut on object type (product / service / MRO), not on decoupling point. D1-D4 are gone outright; there is no D prefix in SCOR DS at all.

The definition of Level 2 changed, and the change is not cosmetic

SCOR 6.0, quoted in the parent brief: "Configuration Level (Process Categories) … A company's supply chain can be 'configured-to-order' at Level 2 from the core 'process categories.' Companies implement their operations strategy through the configuration they choose for their supply chain."

SCOR DS v14, verbatim: "Level-2 process categories determine the capabilities within the level-1 processes. For example, the key level-2 processes for Plan are Plan Supply Chain, Plan Order, Plan Source, Plan Transform, Plan Fulfill, and Plan Return; the level-2 processes for Source are Strategic Source, Direct Procure, Indirect Procure, and Source Return."

The word "configuration" has moved. In v14 it is no longer a property of Level 2 — it is a step in the improvement methodology: the SCOR Racetrack's third step, "Configure the supply chain: Determine the performance metrics and processes of the SCOR improvement program." Level 2's stated job is now capability determination and structural diagnosis ("level-2 process diagrams help identify structural issues in the supply chain, such as why a company has a warehouse feeding a warehouse feeding a warehouse").

The quotation the parent brief hung its structural argument on cannot be cited as current text. The behaviour is still there — you still select which Level-2 categories your chain runs — but the sentence that says so in so many words is gone.

Where D4 went: nowhere. It was demoted, not relocated

The v14 change log, verbatim: "Separated Deliver into Order and Fulfill to better represent omni-channel, transportation, and warehouse management" and "Expanded Make to Transform and updated Return to include services."

The word "retail" appears zero times anywhere in the SCOR DS Quick Reference Guide — not in a Level-1, Level-2 or Level-3 label. There is no Fulfill Retail Product, no retail branch, no retail variant.

What ASCM did instead is stated explicitly, and it is the sharpest finding in this brief. From the front matter, on the boundary between Level 3 and Level 4:

"For example, most if not all companies need to perform a task known as receive order signal. This is a level-3 process (F1.1, F2.1 & F3.1, for example). The level-4 processes would describe the steps to receive the order signal, such as whether the order was received electronically, over the phone or in person. Each of these scenarios may require a unique level-4 process description. … For example, electronic orders may be automatically processed by software, phone orders might be handled by the order desk, and in-person orders would be handled at the checkout counter."

And on Level 4's status: "Companies may develop standard process descriptions for activities within the level-3 processes, which would become level-4 processes. These processes generally are industry-, product-, location- or technology-specific."

So: the in-store checkout transaction — the thing D4 Deliver Retail Product existed to name — is used by ASCM as the canonical example of a Level-4 concern, and Level 4 is the level SCOR deliberately does not standardize. Retail did not get a new code. It got pushed below the floor of the standard.

Its nearest structural home is O1 Order B2C + F1 Fulfill B2C, which is where a store sale now lands, sharing a category with e-commerce home delivery, BOPIS and direct mail-order. The distinction SCOR 6.0 thought was worth a top-level code, SCOR DS thinks is an implementation detail beneath a customer-type category.

Dating the deletion — secondary only, and it stays that way

Secondary sources agree that D1-D4 including D4 Deliver Retail Product persisted from its introduction in SCOR 6.0 through SCOR 12.0 (2017, the last pre-DS release), and that the restructure landed with the SCOR Digital Standard in 2022. The v14 front matter's acknowledgements do thank contributors to "the SCOR DS 2022 update," which is consistent. But the apics.org v12.0 framework PDF is 403-blocked, so the "stable through 12.0" half of this is corroborated only by secondary sources and remains unverified against primary text. Note also that the change log is headed "SCOR v14 Updates" while containing the 2022 Deliver-split item, so it reads as cumulative-since-12.0 rather than strictly v13→v14; treat the per-version attribution as soft.

Convergences and contradictions

• Convergence, and it is the useful one. Vault and primary agree that SCOR runs a two-level architecture: Level 1 is a universal decomposition every chain performs, Level 2 is a smaller set of named alternatives a given company partially instantiates. That architecture survived a total content re-cut — ASCM replaced every Level-2 label and kept the shape. A structure that survives its own contents being thrown out is better evidence for the structure than one that never got tested.
• Contradiction, and it is real. The parent brief wrote that SCOR's Level-2 execution set is "precisely our two axes, named and separated, in a non-GTM function, published by an industry body in 1996 and stable through SCOR 12.0." The stability half is false as stated and false in the direction that matters: it implies currency, and the catalog it describes has been fully superseded. Every SCOR code quoted in that brief — S1-S3, M1-M3, D1-D4 — is a dead code in the live standard, with three of them reassigned to unrelated processes.
• Contradiction on the axis itself. SCOR 6.0's Level-2 axis was fulfillment strategy / decoupling point (stocked · make-to-order · engineer-to-order · retail). SCOR DS's Level-2 axis for Order and Fulfill is customer relationship type (B2C · B2B · intra-company); for Source it is sourcing mode (strategic · direct · indirect); for Transform it is object type (product · service · MRO). Same architecture, four different axes, none of them the old one. Make-to-stock / make-to-order / engineer-to-order does not appear as a Level-2 category anywhere in SCOR DS.

Synthesis for RDCO

Verdict: the structural argument stands, the evidence for it has to be rebuilt, and one specific claim must be withdrawn from the parent brief before anyone quotes it to a client.

What stands. The proposition that a mature, industry-body-maintained reference model separates decomposition (Level 1, everybody does all of these) from selectable alternatives (Level 2, you run a subset) is now verified against the current standard rather than a 2003 booklet, and verified more strongly than before. ASCM re-cut every Level-2 label in the model and preserved the two-level shape anyway. The selectivity test also passes more cleanly on the new catalog than the old: a pure B2B contract manufacturer runs O2/F2 and not O1/F1; a DTC brand runs O1/F1 and not O3/F3; a services firm runs T2 and not T1. "Which of these do you run?" carries information in every row. So the proposed axis + variants amendment — motions-as-choice living one level below sub-functions rather than beside them — keeps its precedent.

What has to be rebuilt. Three things. First, the citation. Every SCOR code in the parent brief is stale, and three of them (S1, S2, S3) now name different processes, which is worse than stale — it is actively misleading to a supply-chain-literate reader, exactly the kind of reader at Kwik Trip. That table needs replacing with the v14 catalog above, or explicitly re-labelled as historical. Second, the "configured-to-order at Level 2" quotation. It is the single most persuasive sentence in the parent brief and it is no longer in the standard; v14 says Level-2 categories "determine the capabilities within the level-1 processes" and moves configuration into the Racetrack methodology. We can still argue configuration-by-selection from behaviour, but we cannot put ASCM's current words behind it. Third, the "stable through SCOR 12.0" claim must come out. It was already flagged unsourced; it is now known false in implication.

What genuinely changes our thinking — and it cuts against SCOR as an authority argument. The parent brief's implicit pitch was "don't cite GTM motion blogs to a client, cite SCOR, because SCOR is the standard." That is still directionally right, but the re-cut adds a caveat we should say out loud: ASCM itself does not treat the Level-2 axis as canonical. It changed the axis, wholesale, within one version transition. What SCOR supplies is the pattern of having a configuration layer, not a durable answer about what that layer should be cut on. If we present the Organizational Platform's motion axis as "the industry-standard shape," a sharp client can point at exactly this transition. The honest and stronger framing is: the two-level architecture is the borrowable, durable thing; the specific axis is engagement-local and expected to move. That framing also happens to defuse the parent brief's unresolved fight with its APQC and BIZBOK siblings over which functions have "true motions" — if the axis is local, the per-function verdict is an empirical question per engagement, not a taxonomy to settle in advance.

The Kwik Trip consequence, which is better news than the question implied. D4 Deliver Retail Product was the parent brief's retail anchor and its "for a convenience-store and fuel retailer, D4 is the motion" line is now a claim about a superseded standard — withdraw it. But SCOR DS's replacement fits Kwik Trip better than D4 ever did. Kwik Trip is not a single-motion retailer; it runs vertically integrated bakery, dairy and commissary production feeding its own stores, plus its own distribution fleet. Under SCOR 6.0 that internal leg had no code — D1-D4 were all customer-facing, and the intra-company flow was invisible. Under SCOR DS it is O3 Order Intra-company + F3 Fulfill Intra-company, running in parallel with O1/F1 for the store-to-consumer leg and T1 Transform Product for the manufacturing side. That is a materially more accurate map of that business, and it names the internal supply chain the old model forced you to hand-wave. The loss of D4 costs us a rhetorical anchor and gains us a better model of the actual client.

Why this is in the vault

This closes the last unresolved primary-source gap behind the Organizational Platform's proposed axis + variants schema amendment, which the platform's README.md holds the 10 function pages and 36 motion pages unwritten pending. It also retires three specific claims in [[2026-08-17-gtm-motion-origins-industry-reference-models]] — the S1-S3/M1-M3/D1-D4 code table, the "configured-to-order at Level 2" quotation, and "stable through SCOR 12.0" — that were on track to be repeated to a Kwik Trip audience where a supply-chain-literate reader would catch them.

Open follow-ups

• Does the SCOR DS full model at scor.ascm.org (behind the free account gate, not opened here) publish an explicit old→new crosswalk table mapping D1-D4 and M1-M3 onto the O/F/T codes, or did ASCM ship the restructure without a migration path for existing SCOR implementations?
• Did make-to-stock / make-to-order / engineer-to-order survive anywhere in SCOR DS at Level 3, as a Best Practice, or as a performance-attribute qualifier — or was the decoupling-point concept dropped from the standard entirely?
• Can the SCOR 12.0 framework introduction be reached through a non-apics.org mirror (the Georgia Tech ISyE mirror served SCOR 6.0), to primary-verify the "D1-D4 persisted through 12.0" claim that currently rests on secondary sources only?
• ASCM re-cut the Level-2 axis from fulfillment strategy to customer type between 12.0 and DS — what was the stated rationale in the 2020-2022 public call for comment, and does it generalize as an argument for why any Level-2 axis is engagement-local rather than canonical?
• The v14 change log reads as cumulative since 12.0 rather than strictly v13→v14; is there a published per-version change log that separates the 2022 SCOR DS release from the 2025 v14 revision?
• SCOR DS added OE13 Circular Supply Chain and sustainability content across every process — does that create a second orthogonal axis (linear vs circular) that the Organizational Platform's variants schema would need to represent, distinct from the customer-type axis?

Related

• [[2026-08-17-gtm-motion-origins-industry-reference-models]] — the parent brief this verifies; its structural spine survives, three of its specific claims do not.
• [[2026-08-21-apqc-pcf-80-channel-motion-leaves]] — the sibling verification from the same round, and the source of the gated-listing-vs-open-asset retrieval trick reused here.
• [[2026-08-17-apqc-pcf-function-motion-mapping]] — sibling brief; its function/motion verdicts are affected by the "axis is engagement-local" conclusion above.
• [[2026-08-17-bizbok-operating-model-function-motion-mapping]] — sibling brief proposing a competing configured-as schema amendment; still unreconciled with axis + variants.
• [[2026-08-04-kwik-trip-ai-architecture-engagement-context]] — the live engagement whose retail anchor moves from D4 to O1/F1 + O3/F3.
• [[01-projects/phdata/2026-06-29-ai-capability-org-structures]] — our own six-pattern delivery-model catalog, an instance of the same configuration-layer genre.

Sources

Primary, fetched and read this session:

• ASCM, SCOR Digital Standard — Quick Reference Guide (full Level-0 → Level-3 process chart). https://www.ascm.org/globalassets/documents--files/corporate-transformation/scor-ds-digital-guide_final.pdf — downloaded, text-extracted locally, read in full. Source of the complete Level-2 code table.
• ASCM, SCOR Digital Standard — Introduction and Front Matter, SCOR Version 14.0, CC BY-NC-ND 4.0, © 2025 ASCM. https://www.ascm.org/globalassets/ascm_website_assets/docs/scor/intro-and-front-matter-scor-digital-standard-2025.pdf — downloaded, text-extracted locally, read in full. Source of the level definitions, the Level-3/Level-4 boundary example, the Racetrack methodology, and the v14 change log.

Primary, blocked:

• APICS, SCOR Version 12.0 Framework Introduction. https://www.apics.org/docs/default-source/scor-training/scor-v12-0-framework-introduction.pdf — HTTP 403 to WebFetch and to browser-UA curl. Not read. This is the document that would primary-verify the D1-D4 persistence-through-12.0 claim.
• ASCM interactive full model, https://scor.ascm.org — free-account gated, not opened.

Secondary:

• Search-result corroboration that D4 Deliver Retail Product was introduced in SCOR 6.0 and persisted through 12.0, and that the Deliver→Order/Fulfill split landed with SCOR DS (2022). Corroborated across multiple independent summaries; no single authoritative secondary source fetched.
• Wikipedia, "Supply chain operations reference" — surfaced in search, used only as a cross-check on the Level-1 list, which is independently primary-verified above.

Vault:

• ~/rdco-vault/06-reference/research/2026-08-17-gtm-motion-origins-industry-reference-models.md — see [[2026-08-17-gtm-motion-origins-industry-reference-models]]
• ~/rdco-vault/06-reference/research/2026-08-21-apqc-pcf-80-channel-motion-leaves.md — see [[2026-08-21-apqc-pcf-80-channel-motion-leaves]]
• ~/rdco-vault/06-reference/research/2026-08-17-apqc-pcf-function-motion-mapping.md — see [[2026-08-17-apqc-pcf-function-motion-mapping]]
• ~/rdco-vault/06-reference/research/2026-08-17-bizbok-operating-model-function-motion-mapping.md — see [[2026-08-17-bizbok-operating-model-function-motion-mapping]]

Confidence: High on the SCOR DS Level-2 catalog, the absence of any retail category, and the Level-4 demotion of in-person order handling — all read directly off two ASCM primary PDFs. High that S1-S3/M1-M3/D1-D4 are absent from the current standard and that S1-S3 are reassigned. Medium on the version at which the deletion occurred, which rests on secondary sources because apics.org is 403-blocked. Medium on the interpretive claim that O1/F1 is D4's nearest structural home — that is my reading of the category definitions, not an ASCM-stated mapping, and the crosswalk follow-up above would settle it.