PCF 8.0 left the channel-motion leaves childless — the orthogonal-axes finding survives

The question

Verbatim: "Did APQC PCF 8.0 give the channel-motion leaves 3.5.6 / 3.5.7 / 3.5.8 (physical-outlet, field, digital sales) any children, or do they remain childless as in 7.4?"

Context: this is the single question that could reverse the core finding of [[2026-08-17-apqc-pcf-function-motion-mapping]] — that APQC named our exact motion axis and then declined to decompose it, which is the evidence that PCF and the Organizational Platform's function/motion tree decompose on orthogonal axes. The parent brief closed with this explicitly unverified, because category 3 sat past the free-preview boundary of every route then available.

Answer: no children. All three remain bare leaves in 8.0, with their 7.4 concept IDs intact. Verified against the full PCF 8.0 PDF, read directly.

What we already know (from the vault)

• [[2026-08-17-apqc-pcf-function-motion-mapping]] established from a direct read of PCF 7.4 that CI 3.5.6 Perform sales at physical outlets (21427), 3.5.7 Perform field sales (21428), and 3.5.8 Perform digital sales (21429) are childless level-3 processes, and built its structural critique on exactly that observation.
• That brief's close-out addendum reached only pages 1–5 of 38 of the 8.0 document via a Scribd OCR preview, and recorded three things as open: whether category 3 changed, whether the category-7 label really drifted from "Human Capital" back to "Human Resources," and whether the 8.0 PDF was obtainable at all without a lead form.
• The dual-identifier discipline (stable 5-digit concept id decoupled from unstable positional path) was named there as the single highest-value thing to steal from PCF. That recommendation is now testable against a real version transition, and it holds up — see below.
• [[2026-04-03-management-is-prediction-spc]] frames the org as a process graph with manipulable inputs, which is PCF's worldview and precisely the axis our tree does not run on.

What the web says

The route that worked — and it closes the parent brief's blocked-download gap

The parent brief's dead ends were real: the resource-listing page for the 8.0 cross-industry PDF is lead-gated behind apqc.org/resource-library/free-content/nojs?nid=…, and apqc.org HTML returns HTTP 403 to automated fetch. I confirmed the same gate on the per-category document as well — the "3.0 Market and Sell Products and Services Definitions and Key Measures PCF version 8.0" listing (published 2026-03-24, node 550870) returns "Please fill out this form to view your content" with First Name / Organization Email fields. Per the no-lead-gen rule I did not submit it. That untried route is therefore blocked, not exhausted-and-negative.

What worked instead: APQC's Drupal static-asset path serves the file unauthenticated when you know the exact filename. Two schemes are live — apqc.org/sites/default/files/<name>.pdf (legacy) and apqc.org/system/files/resource-file/YYYY-MM/<name>.pdf (current). The gate is on the listing page, not on the asset. Filename convention is K<6-digit item code>_<exact resource title>.pdf, and the folder is the publication month of the asset, not of the listing.

The parent brief guessed at this scheme and got 404s because it used the 7.4 file's title styling (Cross Industry, space) rather than the 8.0 listing's (Cross-Industry, hyphen), and did not try the 2026-02 folder. The working URL:

https://www.apqc.org/system/files/resource-file/2026-02/K016809_APQC%20Process%20Classification%20Framework%20%28PCF%29%20-%20Cross-Industry%20-%20PDF%20Version%208.0.pdf

850 KB, 37 pages, HTTP 200, no cookie, no referrer. Full text extracted and read. Note the 404-vs-403 signal that made the search tractable: a wrong filename on a live folder returns 404, a wrong folder returns 403, so probing distinguishes "bad name" from "bad path."

The answer, from the primary source

Cover and running footer: "Version 8.0 • February 2026" / "© 2026 APQC. ALL RIGHTS RESERVED." Category 3 spans pages 8–10; the three leaves close the category at the bottom of page 10's right column, immediately before the footer and category 4.0 on page 11. Verbatim, as rendered:

3.5.6 Perform sales at physical outlets (21427) 3.5.7 Perform field sales (21428) 3.5.8 Perform digital sales (21429)

Three verification passes, all agreeing:

1. Regex sweep of the extracted text for 3.5.6.x / 3.5.7.x / 3.5.8.x anywhere in the 37-page document: zero matches. Same sweep on 7.4: zero matches. No child element exists under any of the three in either version.
2. Page image read. Page 10 was rendered at 150 dpi and read as an image, because the parent brief correctly warned that PCF's graphics do not OCR and text extraction can drop content. The image confirms the three leaves as flush level-3 entries with no indented children beneath them, and confirms 3.5.8 is the last element in category 3 — there is no 3.5.9.
3. Concept-ID continuity. All three carry their 7.4 IDs unchanged (21427 / 21428 / 21429). By APQC's own stated rule — "A new 5-digit number will be assigned to a process element if its definition substantially changes" — retained IDs are APQC asserting these definitions did not substantially change.

Category 3 did change — just not where it would have mattered

The null result on 3.5.6–3.5.8 is not because 8.0 left category 3 alone. It didn't:

• 3.2 Develop marketing strategy was re-minted: ID 10102 (7.4) → 21615 (8.0). Same name, new concept id, which is APQC declaring a substantial definitional change.
• 3.2 gained a process group, going from six to seven. 8.0's list: 3.2.1 Define offering and customer value proposition (11168) · 3.2.2 Develop and manage brands (11445) · 3.2.3 Define pricing strategy (10123) · 3.2.4 Define and manage channel strategy (20000) · 3.2.5 Analyze and manage channel performance (20006) · 3.2.6 Develop marketing communication strategy (16848) · 3.2.7 Design and manage customer loyalty program (18924). In 7.4 the same trailing five sat one position earlier (pricing at 3.2.2, channel strategy at 3.2.3, and so on).
• A new node landed inside channel strategy: 3.2.4.9 Define and manage channel/store format strategy (11458). 7.4's channel-strategy group stopped at .8.
• Neither 11445 nor 11458 appears anywhere in 7.4. Their low ID range suggests reinstated pre-7.4 elements rather than freshly minted ones; I did not verify that against a pre-7.4 edition, so treat "absent from 7.4, present in 8.0" as the verified claim and the reinstatement story as inference.
• 3.4 Develop sales strategy (10103) and 3.5 Develop and manage sales plans (10105) kept both name and id.
• Whole-document scale: 8.0 carries roughly 2,030 5-digit-id occurrences against 7.4's 1,933, about 5% growth. APQC was actively editing. It edited around the channel leaves.

The renumbering is itself the best available evidence for the parent brief's id recommendation. Channel strategy moved from 3.2.3 to 3.2.4 across one minor-to-major version bump while keeping concept id 20000. Any consumer that had keyed on the dotted path silently broke; any consumer that keyed on 20000 did not. This is not a hypothetical failure mode we are guarding against — it fired, in the one framework we were considering borrowing from, in the one version transition we examined.

Two open items from the parent brief, now closed against primary source

• Category 7 rename: confirmed. 8.0's table of contents and category header both read 7.0 Develop and Manage Human Resources (10007), where 7.4 reads "Human Capital." The label did drift backward to the older wording. The concept id 10007 is unchanged, so APQC treats it as a rename, not a redefinition. The parent brief's "worth a second look before it is quoted to a client" caveat can be lifted.
• License grant: confirmed in the full 8.0 document, not just the OCR'd preview. The ©2026 copyright page carries the identical "perpetual, worldwide, royalty-free license to use, copy, publish, modify, and create derivative works of the PCF" text, conditioned on the same attribution paragraph. Page footers still read "Permission granted to photocopy for personal use." The 13-category structure, the operating/management split, and the numbering-scheme paragraph all carry forward verbatim.

Convergences and contradictions

• Convergence, and it is the finding. The vault's 7.4-based claim and the 8.0 primary source agree exactly: three named channel-execution processes, zero children, in both versions, eighteen months and one major version apart. The parent brief's observation was not a snapshot artifact of one edition.
• Convergence with a twist that sharpens the argument. APQC did decompose channel work in 8.0 — nine children under 3.2.4 Define and manage channel strategy, including a new store-format node. So APQC is not indifferent to channels. It decomposes channel strategy (a planning process with real steps: determine channels, set objectives, assess fit, select for segments, build a roadmap) and declines to decompose channel execution. That is a coherent editorial position, not an oversight, and it is a stronger version of the parent's argument than the parent made.
• No contradiction found. I looked for one. A vocabulary sweep across both versions for inside sales, e-commerce, direct sales, telesales, self-service, retail store returns zero hits in either. omni-channel appears three times in both, all under channel strategy. 8.0 introduced no operating-pattern vocabulary anywhere.

Synthesis for RDCO

The parent brief's core finding stands, verified against primary source. It is not reversed, and it is no longer unverified. APQC PCF 8.0 (February 2026) leaves 3.5.6 / 3.5.7 / 3.5.8 exactly as 7.4 left them: three childless leaves carrying their original concept ids. The Organizational Platform's content template can lock with the orthogonal-axes argument load-bearing, and the borrow-vs-build verdict table in [[2026-08-17-apqc-pcf-function-motion-mapping]] needs no revision on this axis.

What changed is the quality of the argument, and it got better. A reader could have dismissed the 7.4 observation as APQC simply not having gotten to it yet — three leaves added late in a version cycle, children pending. Eighteen months and one major version later, with roughly a hundred new elements added framework-wide and category 3 specifically reworked — a re-minted 3.2, a new process group, a renumbered channel-strategy branch, a new store-format node — APQC still did not put anything under them. The editors were in that neighborhood, editing. They decomposed the planning of channels nine ways and the execution of channels zero ways. That is a decision, and it is the same decision our tree makes in reverse: we decompose execution by operating pattern because that is where cost structure, headcount shape, and hiring profile live, and those are the facts our 80 role pages hang off. A process taxonomy has nothing process-shaped to say about the difference between a field rep and a store associate closing the same order, so it says nothing. Our tree exists because that silence is expensive.

The second-order finding is the one I would actually act on this week. The dual-identifier recommendation stopped being a design principle and became an observed failure. Define and manage channel strategy moved from 3.2.3 to 3.2.4 between 7.4 and 8.0 while keeping concept id 20000. Any client YAML, mapping table, or deliverable that cited the dotted path is now silently wrong; anything citing 20000 is still correct. Our current ids (FN.SALES.INSIDE) are positional, semantic, and stable all at once, which is three jobs one string cannot hold — the parent brief said so, and PCF just demonstrated the breakage live. Mint stable ids before the first motion is re-parented, not after.

Third, a reusable operational finding worth more than this one question. APQC's lead gate is on the listing page, not on the asset. The static path apqc.org/system/files/resource-file/YYYY-MM/K<code>_<exact title>.pdf serves unauthenticated, and the 404-vs-403 distinction makes filename probing tractable without brute force. Every remaining "APQC content we could not reach" item — the per-category Definitions and Key Measures documents, industry-specific PCFs, the 8.0 Excel — is likely reachable the same way, given the item code and the exact resource title from the listing page. That converts a whole class of previously-blocked APQC questions into solvable ones, and it does it without registering for anything.

Why this is in the vault

This closes the one reversal condition that was holding the Organizational Platform's phase-3 content template open. The hand-section template for the 10 function pages and 36 motion pages can now lock with the "motion layer is built, not borrowed" verdict resting on a verified primary source rather than a single-version observation, and the platform's id scheme should be changed to stable-concept-ids before the first motion re-parenting rather than after — a decision this brief moves from "recommended" to "demonstrated necessary."

Open follow-ups

• The APQC static-asset bypass should be tested against the per-category Definitions and Key Measures documents for 8.0. Category 3's would give us APQC's prose definitions of 21427 / 21428 / 21429 — the sentences describing what "perform field sales" means to APQC. That is different information from the hierarchy, and it may reveal whether APQC intends them as processes or as tagging dimensions. Blocker is only the item code; the listing page for each carries the node id.
• Does APQC publish a 7.4 → 8.0 change log or release note? The structural diff in this brief was reconstructed by comparing two PDFs, which catches renumbering and additions but would miss deletions I did not think to look for. An authoritative change list would let us verify the diff rather than trust it.
• 3.2 Develop marketing strategy was re-minted from 10102 to 21615 — APQC's signal that the definition substantially changed. What actually changed in the definition, and does it affect the FN.MARKETING borrow-vs-build verdict? The definitions document would answer this.
• The industry-specific PCFs are being re-cut at 8.0 (a Utilities Excel v8.0 appeared 2026-08-20). Is a Retail PCF 8.0 in that program, and if so does it decompose store-channel execution the way cross-industry declines to? Retail is the one industry where APQC has a business reason to. If it does, the orthogonality argument needs a stated exception for retail-format clients — which is Kwik Trip.
• Do 11445 (Develop and manage brands) and 11458 (channel/store format strategy) exist in pre-7.4 cross-industry editions? If they are reinstatements rather than new elements, PCF's concept-id space is being reused across gaps, which is a wrinkle in the id scheme we are about to copy.

Related

• [[2026-08-17-apqc-pcf-function-motion-mapping]] — the parent brief this verifies; its core finding stands and two of its flagged uncertainties are closed here.
• [[2026-08-17-bizbok-operating-model-function-motion-mapping]] — sibling from the same research round; concluded the founder's original prior on the motion axis survives, which this brief independently corroborates from the PCF side.
• [[2026-08-17-gtm-motion-origins-industry-reference-models]] — sibling that graded Marketing and IT as sub-functions rather than true motions, disagreeing with the parent. Carries a VERIFICATION STATUS: ITERATE (hard) banner; nothing here resolves that disagreement.
• [[2026-08-04-kwik-trip-ai-architecture-engagement-context]] — the live engagement; the Retail PCF 8.0 follow-up above matters specifically for it.
• [[2026-04-03-management-is-prediction-spc]] — the process-as-manipulable-system worldview PCF embodies and our tree deliberately does not.

Sources

Primary, read directly (full text + page-image verification):

• APQC's Process Classification Framework® — Cross-Industry, Version 8.0 (cover/footer: "Version 8.0 • February 2026"; ©2026 APQC), APQC item code K016809, 37 pages. Direct unauthenticated static asset: https://www.apqc.org/system/files/resource-file/2026-02/K016809_APQC%20Process%20Classification%20Framework%20%28PCF%29%20-%20Cross-Industry%20-%20PDF%20Version%208.0.pdf — full text extracted; page 10 additionally rendered at 150 dpi and read as an image to confirm the absence of child elements under 3.5.6 / 3.5.7 / 3.5.8.
• APQC's Process Classification Framework® — Cross-Industry, Version 7.4 (August 2024), item code K014750, public mirror at https://solutions.ifrc.org/sites/default/files/2024-10/K014750_APQC%20Process%20Classification%20Framework%20(PCF)%20-%20Cross%20Industry%20-%20PDF%20Version%207.4.pdf — re-downloaded and diffed against 8.0 for this brief.

Blocked (lead-gen form; not submitted, per standing rule):

• 3.0 Market and Sell Products and Services Definitions and Key Measures PCF version 8.0 (published 2026-03-24, node 550870): https://www.apqc.org/resource-library/resource-listing/30-market-and-sell-products-and-services-definitions-and-key-3 — "Public Content: View Now" resolves to apqc.org/resource-library/free-content/nojs?nid=550870, which serves a First Name / Organization Email form. Not attempted further. Item code unknown, which is the only thing preventing the static-asset bypass from reaching it.
• PCF Version 8.0 Process Definitions and Key Measures Collection: https://www.apqc.org/resource-library/resource-collection/pcf-version-80-process-definitions-and-key-measures-collection — index read via readability proxy; all thirteen per-category documents behind the same gate.

Vault:

• rdco-vault/06-reference/research/2026-08-17-apqc-pcf-function-motion-mapping.md
• rdco-vault/06-reference/research/2026-08-17-bizbok-operating-model-function-motion-mapping.md
• rdco-vault/06-reference/research/2026-08-17-gtm-motion-origins-industry-reference-models.md
• rdco-vault/01-projects/phdata/2026-08-04-kwik-trip-ai-architecture-engagement-context.md
• rdco-vault/06-reference/2026-04-03-management-is-prediction-spc.md