Operating Principles for Work Placement
Part of the Planning Week chain · previous: Strategy on a Page · next: Decision Rights by Planning Horizon
Operating principles for work placement are the twelve rules of the road a workforce function adopts when it takes on the placement of work as a standing function, each written on a card with a focus area, the rule, a rationale and a statement of what will be different. They matter because a principle is the cheapest decision a function can make: agreed once, it settles every later argument of its kind without a meeting. The page produces the principle cards (PC) and is worked in a session with Wiki:Packs/Strategy on a Page (CP-WFM-011); the cards are ratified with the strategy as decision D-01 on Day 1 morning of a planning week.
The card format
Each principle is one card with four fields. The format is kept spare because a rule without a stated consequence is not kept; the content is written for a function whose defining question is where work sits.
| Field | What it holds | Why it is there |
|---|---|---|
| Focus | The area the rule governs: placement, decisions, data, process, value, measurement, maturity, technology | So a card can be found when its kind of argument arrives |
| Rule of the road | The principle as an instruction, in one sentence, with no qualifier | A principle with a qualifier is a policy, and policies are negotiated at the point of decision |
| Rationale | Why the rule holds, in two or three sentences, with what it rests on named | So the rule can be re-examined when what it rests on changes |
| What will be different | The observable change once the rule is kept: what stops being said, what starts being recorded | The test of adoption; a principle no one can see kept is not adopted |
The fourth field is the one that earns the card its place. A room can agree a rule in a minute; it agrees what will be different only after it has imagined the rule applied to the argument it had last month. Simple rules of this kind are how organizations act coherently in conditions where a detailed plan would not survive contact,[1] and the fourth field is where a rule proves it is simple enough to keep.
The twelve principles
The generic set for a function moving toward a global ROC, the shared service that owns the methodology for forecasting, scheduling, real-time steering and work placement across every node. A node is a place work can sit: a hub (in-country depth), a service center (badged arbitrage at scale), a partner (speed, market access, flex), or automated. The rationale column names the page that carries the argument; this page does not restate those arguments.
| # | Focus | Rule of the road | Rests on | What will be different |
|---|---|---|---|---|
| 1 | Placement | Work sits where objectives and constraints say, not where it has always sat | Sourcing Design Axes: Node and Client Ownership | "History" stops being a reason; it becomes a recorded constraint or it is dropped |
| 2 | Placement | A hard constraint excludes; it never lowers a score | Placement Engine Architecture | Badging, regulation and contract terms stop being argued as points; they are tested before scoring |
| 3 | Placement | Parity at matched tenure makes work a candidate; a move must outweigh the re-ramp, this chain's term for the cost of rebuilding proficiency the move gives up | Right Work, Right Capability, Right Location (capability read at matched tenure, never at headcount) · Placement Rules and the Tenure Contract (the tenure contract the re-ramp prices) | Every gate run carries the re-ramp as a priced line |
| 4 | Placement | A partner placement carries one supply-seat use, never mixed in one pool: valve (variance absorbed), held (base load at parity or for market access), or automating (work leaving the post, which shifts to oversight) | The Vendor as Relief Valve · Outsourcing as a Risk Lever · see the reconciliation below on why the card's three uses are not the two commercial justifications | Each supply-seat card says which; the contract form follows the use |
| 5 | Decisions | The function owns the methodology; the node owner owns the decision; every decline is recorded with its price | Right Work, Right Capability, Right Location (the rule, stated there) · Vendor Governance Placement (standard-setting, measurement and enforcement are separable) · The Workforce Broker | No placement is "the planning team's call"; no decline is free of a record |
| 6 | Data | One definition, one core, one instrument; build the standard concurrently with the climb, never serialized behind it | Platform Migration as a Definitional Forcing Function (definitions become critical path when a migration is committed) · Vendor Ecosystem Restructuring Agenda (one instrument and one taxonomy run first, and in parallel) | No program waits for the definitions register; the register runs beside it |
| 7 | Process | Written-down work goes to agents with an overseer and a catch rate; uncertain work stays with people who return a range | Three Bands of Work · The Agent Overseer | "A range, not a number" governs the function's own outputs |
| 8 | Value | Value has two sides, commercial and interaction, held side by side and never blended | Value First, Then Route · Quality Score and Customer Experience Index | The CX index and the quality score are paired on one card, never averaged |
| 9 | Value | Queueing arithmetic sizes the pool after the work is classified, never before | Erlang C · Value First, Then Route | The queue stops defining the pool; value and capability do |
| 10 | Measurement | A finding about the instrument is never a finding about the person | Sample Size and Detectable Difference in Quality Measurement (a gap below the detectable difference is a fact about the sample, not about the unit) | Instrument effects are reported before performance is |
| 11 | Maturity | A level is held only when the level below is satisfied; aspiration is not position | Interpreting WFM Maturity Assessments | An assessment's blended score is never quoted as the level |
| 12 | Technology | Ontology on paper before platform; exhaust the core before buying a tool | Technology Journey from Level 2 to Level 5 | The capability platform is the last purchase, not the first |
Principle 4's count needs a word of reconciliation, because The Vendor as Relief Valve and Vendor Ecosystem Restructuring Agenda both state that a supplier placement carries one of two justifications, variance absorption or market access, and this card names three uses. The two are the commercial justifications for buying a partner node at all. The three are the supply-seat uses the card that governs a block must carry, and they are not the same list: market access is why an estate holds a partner block rather than building a service center, so on the card it appears as a reason for held rather than as a use of its own, and automating is a third state the commercial justifications do not describe, because a block whose work is leaving the post is on a different trajectory from one absorbing variance or holding a market. Those two pages govern the commercial case; this card governs the block. A room that prefers to keep market access as a use of its own amends the card and records it, because the contract form follows the use. One caution the valve page attaches to the held use: base load placed with a partner for rate reasons where a service center exists is a cost decision rather than a flexibility one, and it competes on total cost with control or it does not move.
Two things are absent on purpose. There is no principle that says "always keep high-value customers in-house," because that is the rate-card thinking principles 1 to 3 replace: value is classified first and location follows. And there is no separate vendor-management focus area, because the partner is a node inside principle 1, a party inside principle 5, a supply-seat use inside principle 4 and a pool inside principle 9; a vendor principle of its own would produce the second, weaker operating model that Operating Model for Workforce Management warns against.
Three of the twelve rest on old results. Queueing formulas size a pool from arrivals, handle time and a service target and say nothing about which work should be in the pool,[2] so principle 9 fixes the arithmetic's place in the order rather than demoting it. An operational definition is what makes a number mean the same thing to two people,[3] and principle 6 adds when they are built: beside the climb. And a measure applied to people reshapes the behavior it measures,[4] which is why principle 10 reports the instrument's effect before anyone's performance.
Where "level" appears on this page it means the maturity model's five levels, except in principle 7, where "written down" means the L0–L3 documentation levels of Process Decomposition (L0–L3).
The three that break something if struck
A room may amend the wording of any card and strike most of them without damaging the model. Three are structural: each holds up a later page of the chain, and striking it does not weaken that page but removes it.
| Principle | What it holds up | What happens without it |
|---|---|---|
| 2 A hard constraint excludes; it never lowers a score | The Placement Gate | A constraint that can be outweighed by a score is a soft constraint; the gate's fail outcome disappears and every exclusion becomes a negotiation at the point of decision |
| 5 The function owns the methodology; the node owner owns the decision | Decision Rights by Planning Horizon | The RACI's first rule has no principle to rest on; either the function becomes the decision-maker it is not staffed or authorized to be, or the node owners own the method and it fragments by owner |
| 8 Value has two sides, never blended | Quality Score and Customer Experience Index and the scorecard | The CX index becomes a weight inside a quality score, the commercial side of value disappears into the interaction side, and the value unit the scorecard reads cannot be defined |
These three are the only live argument on the page, and the facilitator says so before the cards are read. Everything else is wording.
How a room adopts or amends
The cards are read in order, one to a slide or one to a sheet, and the room may do three things with each: adopt, amend the wording, or strike. Adoption is by silence: a card stands when no one wants it removed. An amendment is written on the card in the room, and the amended card is what the record carries. A strike is recorded with its reason, and if the struck card is one of the three above, the record also names the later page the strike affects, so the room strikes with its eyes open.
Three rules keep the block to its time. A card is not argued from its rationale; the rationale is on the page it names, and the argument goes there after the week. A card's what will be different may be rewritten freely, because that field is the room's own. And the cards are ratified with the strategy as one act, D-01 on Strategy on a Page, not twelve votes; a room that wants twelve votes does not yet agree the strategy, which is worth knowing on Day 1 rather than Day 4.
The cards are then the enterprise row of the principles section in The Function Service Pack: each function owner inherits the twelve as written and adds their own what will be different beneath each, which is how a principle agreed by a leadership room reaches the people who keep it.
Worked example
The chain's worked example function adopted all twelve cards on Day 1 morning, Mon 20 Apr 2026, with two amendments and no strikes. Principle 4's rule was amended to name the three supply-seat uses in the order the estate's supply-seat cards would carry them, "valve · held · automating," and its what will be different was rewritten by the room as "every partner block on the org chart carries one word from that list, and a block that cannot carry one is a consolidation candidate." Principle 7's what will be different was rewritten as "the daily note from the agent team on the migrating book carries a catch rate before it carries a recommendation," which the room wrote against the rung the agent team on that book had already reached, pilot, with the planner gate on (The Agent Team Ladder: Alpha to Production). The three structural principles were named aloud and none was contested; the record for D-01 carries "twelve adopted, two amended, none struck." In the validation circuit closing Fri 8 May 2026, one director asked whether principle 3's re-ramp line should carry a number; the answer recorded was that it carries a range with its grade, per principle 7, and the worked figures in the series are the book's 452 s [M] handle time against the 412 s [A] carried from the sending platform, and, beneath it, a partner cohort flat at 470 s [M] against an in-house cohort falling from 430 s to 405 s [M] (Hypothesis Testing with Agent Teams) — a difference the settling test of question Q-004 recorded as structural at Inferred rather than as a re-ramp.
The artifact this page produces
The principle cards (PC), twelve, one per principle, four fields each. They travel with the decision record for D-01 and become the enterprise row of the service pack's principles section. One filled example card:
| # | Focus | Rule of the road | Rationale (names the page) | What will be different (the room's words) |
|---|---|---|---|---|
| 4 | Placement | A partner placement carries one supply-seat use, never mixed in one pool: valve · held · automating | The valve absorbs variance the two fixed nodes should not hold; a held placement is base load kept at parity or for market access; an automating placement is work leaving the post (The Vendor as Relief Valve) | Every partner block on the org chart carries one word from that list, and a block that cannot carry one is a consolidation candidate |
Produced in a working session with Wiki:Packs/Strategy on a Page (CP-WFM-011); the filled set is part of blueprint v0.1.
What would change this
The page's claim is that twelve cards, three of them structural, are enough to settle the recurring arguments of a placing function without a meeting. The observation that would overturn it is a decision register, six months after the week, in which the same kind of argument recurs under a principle the room adopted, and the record shows the principle was cited and overridden rather than not known. That would mean the card's what will be different was not observable enough to be enforced, and the fix is not a thirteenth card but a rewrite of the fourth field on the one that was overridden.
How this connects
- Previous in the chain: Strategy on a Page — the cards are ratified with it as D-01
- Next in the chain: Decision Rights by Planning Horizon — the RACI whose first rule is principle 5
- Defers to: Placement Rules and the Tenure Contract (the discriminating question; the tenure contract the re-ramp prices) · Sourcing Design Axes: Node and Client Ownership (why location is a consequence) · The Vendor as Relief Valve (the valve and the held placement) · Quality Score and Customer Experience Index (paired, never blended) · Erlang C (the arithmetic principle 9 orders)
- Feeds: The Placement Gate (principle 2 is its fail outcome) · Work Placement Landscape (principles 1 and 4 are its transitions) · The Function Service Pack (the enterprise row of section 5)
Maturity Model Position
Written principles are the first artifact of the Level 3 Progressive level on the WFM Labs Maturity Model™: a Level 2 Foundational function runs on inherited rules and a Level 3 Progressive one on adopted ones (Level 2: Structured Workforce Management, Level 3: The Automation Layer). Principles 7 and 12 describe Level 4 Advanced practice and are adopted before it is reached, which is how a Level 3 function keeps its Level 4 purchases in order. Four scales on this wiki use the word "level"; Planning Week for a Workforce Function states which is which. This page uses the maturity Levels 1–5, except in principle 7, where it uses the L0–L3 documentation levels.
See Also
- Planning Week for a Workforce Function
- Operating Model for Workforce Management — the component the cards answer
- Right Work, Right Capability, Right Location — the vision the twelve serve
- Interconnected Workforce Management — where principle 5 meets the other departments
References
- ↑ Eisenhardt, K. M., & Sull, D. N. (2001). "Strategy as Simple Rules". Harvard Business Review 79(1), 106–116.
- ↑ Gans, N., Koole, G., & Mandelbaum, A. (2003). "Telephone Call Centers: Tutorial, Review, and Research Prospects". Manufacturing & Service Operations Management 5(2), 79–141. doi:10.1287/msom.5.2.79.16071.
- ↑ Deming, W. E. (1986). Out of the Crisis. Cambridge, MA: MIT Center for Advanced Engineering Study. Chapter 9, "Operational Definitions, Conformance, Performance". ISBN 978-0-911379-01-3.
- ↑ Ridgway, V. F. (1956). "Dysfunctional Consequences of Performance Measurements". Administrative Science Quarterly 1(2), 240–247. doi:10.2307/2390989.
