Contradiction Resolution (Ideal Final Result)

Quick facts

Mechanism: Converts a stuck trade-off into a contradiction to dissolve rather than a compromise to settle for.

Produces: contradiction-resolution-worksheet

Evidence: M/P · Domain: Problem Framing

Use when a decision has hardened into an either/or like speed versus quality, thorough versus fast, or generous versus profitable, when splitting the difference feels wrong and a win-win would be worth far more than the best compromise, or when a recurring design or product trade-off keeps forcing the same painful balance.

Most problems arrive as a trade-off: to get more of A you must accept less of B (cheaper but flimsier, faster but riskier, generous but unprofitable). The reflex is to optimize - find the least-bad point on the curve. Contradiction resolution refuses that reflex first. It reframes the problem as a contradiction to be dissolved - a solution in which you stop trading A against B at all - and only settles for the compromise once a disciplined attempt to dissolve the trade-off has actually failed. The reframe is the durable move; it is the de-branded core of TRIZ (Genrich Altshuller, from 1946). The output is a contradiction-resolution worksheet, not a discussion.

When to Use

• A problem is stated as, or has quietly collapsed into, an “A vs B” trade-off, and nobody has tested whether the trade-off is actually necessary.
• The obvious answer is “pick a point on the curve,” and a genuine win-win would be worth far more than the best compromise.
• A design, process, or product tension keeps forcing the same painful balance (“thorough vs fast”, “generous vs profitable”, “secure vs convenient”).
• The opposing requirements look like they might hold at different times, places, scales, or under different conditions.

When NOT to Use

• The trade-off is genuinely fundamental. Some contradictions are imposed by physical law, conservation, regulation, or hard external limits and cannot be engineered away (you cannot separate “spend the money” from “keep the money”). Forcing the move there manufactures clever-sounding non-solutions and delays the real decision. When dissolution fails, the honest output is “this is a real trade-off” - route it to think-decision-option-review to choose under it, or treat it as a standing polarity to manage. This is the central wall.
• You need idea volume, not a frame. This skill is convergent on one tension and produces fewer ideas than open ideation. For breadth use think-brainwriting or think-far-analogy-ideation.
• The tension is one you must manage, not eliminate. An unresolvable polarity (centralize vs decentralize, structure vs flexibility) is to be navigated over time, not dissolved - that is tension/polarity mapping, not this.
• No real opposition exists. If the problem is vague or under-specified rather than genuinely two-sided, this tool will invent a contradiction that is not there. Frame it first with think-problem-restatement, then run this once a real trade-off is on the table.

Instructions

When asked to resolve a trade-off or break a stuck compromise, follow these steps:

1. Confirm a real trade-off and name the opposing pair. State the problem as an explicit “to get more of A, we must accept less of B.” If there is no genuine two-sided tension, stop and send it to framing (think-problem-restatement). Then classify the contradiction: a technical contradiction (improving parameter A degrades a different parameter B) or a physical contradiction (one single parameter must hold two opposite values at once). The form selects the resolution menu, so name it explicitly.
2. State the Ideal Final Result (IFR). Describe the end-state implementation-free: the required benefit delivered with no new cost, no new harm, and ideally no new mechanism - “the function happens by itself.” Strip every “how.” The IFR is not a goal restatement; stripping the apparatus exposes how much of it is actually necessary and counters the inertia that makes the trade-off feel inevitable.
3. Attempt to dissolve the contradiction. For a physical contradiction, run the four separation principles in turn and record each attempt: separate the opposing requirements in time (each holds at a different moment), in space (each holds in a different place or part), by scale / system level (move a requirement to a super- or sub-system), or by condition (each holds under a different case, segment, or trigger). For a technical contradiction, prompt with the inventive principles (segmentation, asymmetry, nesting, prior action, “the other way round”, and so on). Treat any 39x39 contradiction matrix as a heuristic idea-prompt, never an authoritative lookup.
4. Resolve or exit honestly. If a separation or principle dissolves the contradiction, state the resolved solution concretely - what changes so you no longer trade A against B. If nothing dissolves it, declare the trade-off genuine and route it onward: to think-decision-option-review to choose under it, or to tension/polarity mapping to manage it. Do not manufacture a clever non-solution; the honest exit is part of the method, not a failure of it.
5. Emit the contradiction-resolution worksheet. Produce the artifact in references/TEMPLATE.md: the opposing pair, the contradiction type, the IFR, each dissolution attempt with the operator it used, and the outcome (a concrete resolution, or an honest “real trade-off, routed to …”). The template’s pre-printed evidence caveat is part of the artifact; carry it through verbatim.

Output Format

Use the template in references/TEMPLATE.md. The deliverable is the filled worksheet - opposing pair, IFR, dissolution attempts, and a resolution-or-honest-exit - not a prose essay.

Quality Checklist

Before finalizing, verify:

• The trade-off is written as an explicit opposing pair, and classified as technical (two parameters) or physical (one parameter, two opposite values).
• The IFR is stated implementation-free (no “how”), not a reworded goal.
• Dissolution was attempted with named operators - the four separation principles for a physical contradiction, inventive-principle prompts for a technical one - and each attempt is recorded, not just the winner.
• Any contradiction matrix was used as a heuristic prompt, not an authoritative lookup.
• The outcome is either a concrete resolution OR an honest “this is a real trade-off” with an onward route - never a vague compromise dressed up as a dissolution.
• The output is the worksheet artifact, not prose.
• No overclaiming: the evidence is practitioner-grade and transferred, and the matrix is contested; claim “tests whether the trade-off is real and often dissolves it,” not a measured gain in outcomes (see evidence/dossier.md).

Evidence

Tier P (governing; honest read M/P, capped at P). Structured contradiction-based ideation has real controlled comparison behind it - quasi-experimental studies report TRIZ-style methods improve the novelty and variety of ideas while reducing their quantity (the M-leaning half) - but three deductions hold the grade at P: the studies measure the broad method, not this specific dissolve-plus-IFR move; the signature contradiction matrix is empirically contested (Spreafico and Russo, 2018, find it covers only ~10-15% of problems); and the one AI-context study (AI EDAM, n=32) found brainstorming higher on fluency and originality, with TRIZ ahead only on elaboration and flexibility. All of it is transferred from human practice, not validated on agents. Full grading, sources, and caveats: evidence/dossier.md.

Examples

See references/EXAMPLE.md for a completed contradiction-resolution worksheet on a real decision.

Deep dive: worked example

A full worked run (the shared Northwind scenario)

Contradiction-Resolution Worksheet - Worked Example

A completed run of the contradiction-resolution skill on a real, consequential decision. This is the quality bar a generated worksheet should meet.

Uses the shared recurring scenario (Northwind, a B2B SaaS weighing a self-serve free-tier launch) so examples across skills read as one coherent product. Where abstraction-laddering relocated what altitude to work the free-tier problem at, this skill takes the trade-off that surfaces once a free tier is on the table and tests whether it can be dissolved. See docs/internal/AUTHORING.md.

---

Problem under resolution

• Problem as given: “The free tier has to be generous enough to actually drive self-serve adoption, but limited enough that it doesn’t cannibalize paid conversion. We keep arguing about where to draw the line.”
• The user’s actual goal: Grow qualified pipeline and product advocacy from the free tier without eroding paid revenue.

Summary (top of the artifact)

The team framed this as one dial - “how generous is the free tier” - to be set at a compromise point, where every step toward adoption is a step away from conversion. That is a physical contradiction: generosity must be both high (for adoption) and low (for conversion). Stating the Ideal Final Result (“the free tier sells the paid tier by itself, at near-zero cost”) and running the separation principles dissolves it: separate by condition / scale. Make the free tier fully generous on individual-value features (which drive adoption and advocacy) while gating the team, admin, and scale features that define paid value. Generosity and limitation then apply to different feature axes, so they stop trading against each other - there is no single dial to compromise. (Had Northwind’s product delivered value only at team scale, no such split would exist and the worksheet would have declared a real trade-off; see the exit note.)

The contradiction

• Opposing pair (A vs B): to get more of A: free-tier adoption and advocacy (argues for a generous free tier) we must accept less of B: paid conversion and protected revenue (argues for a limited free tier).
• Type: Physical. A single parameter - the generosity of the free tier - is required to be both high (to win A) and low (to win B). That single-parameter, two-opposite-values shape is what makes the separation principles the right menu.
• Why it has felt inevitable: everyone pictured generosity as one global slider on the whole product, so more of it necessarily meant giving away more of what people pay for.

Ideal Final Result (IFR)

Every prospect who could get value from Northwind reaches full first value for free and becomes an advocate, at near-zero acquisition cost, with zero loss of paid conversion and no new sales effort - the free experience does the qualifying and the selling by itself. (Note there is no “how” here: no plan limits, no trial length, no feature gates - just the end-state, so the apparatus we assumed is exposed as optional.)

Dissolution attempts

Operator	Attempt (how the requirements might be separated)	Result
Separate in time	Generous for a window, then limited: a 14-day full-feature trial that downgrades.	Partial - recovers conversion pressure but loses the always-on, top-of-funnel advocacy a free tier exists for; it becomes a trial, not a free tier.
Separate in space	Generous in a sandbox / sample-data workspace, limited on real production data.	Partial - good for evaluation, weak for genuine adoption and advocacy (people advocate for tools they really use, not demos).
Separate by scale / system level	Generous at single-user scale; limited at team / org scale (seats, volume).	Dissolves - solo usage is exactly where adoption and advocacy form; scale is exactly what enterprises pay for.
Separate by condition	Generous for individual-value use cases; gate the collaboration, admin/SSO, and scale features that define paid value.	Dissolves - generosity and limitation now apply to different feature axes, not one slider.

Outcome

• Resolution (dissolved): Separate by condition and scale. Ship a free tier that is fully generous on individual-value features (the complete single-user workflow, real data, no time limit - so it drives adoption and creates advocates) while gating the features that define paid value: team collaboration, admin/SSO/governance, and scale (seats, volume, throughput). The “generous vs limited” trade-off dissolves because the two requirements no longer touch the same parameter - you are generous on the adoption axis and limited on the monetization axis at the same time. The free tier now grows the funnel and demonstrates exactly what the paid tier adds, which is what makes it sell the paid tier “by itself.” The earlier argument (“where do we draw the one line?”) was solving a contradiction that did not have to exist.

  Exit note (when this would NOT dissolve): the resolution depends on Northwind having real value at the individual scale. If the product only delivered value once a whole team was on it (value emerges at collaboration, not solo), there would be no individual-vs-team axis to separate on, and none of the four principles would split the generosity dial. The worksheet would then declare a genuine trade-off between free-tier reach and paid revenue and route it to think-decision-option-review to choose a deliberate point under it - rather than manufacturing a separation that the product cannot support.

---

Note how the value is in refusing the compromise first: the problem arrived as “set the generosity dial,” and an unaided pass would have proposed a middle setting (some features free, some paid, argue over which). The skill named the contradiction, stated an implementation-free ideal, and ran the separation menu - turning a one-dial compromise into a two-axis design where adoption and monetization stop competing.

Grounding: the full evidence dossier

What the research does and does not show, with graded sources

Evidence Dossier: Contradiction Resolution (Ideal Final Result)

The single source of truth for the contradiction-resolution skill. The SKILL.md, the sidecar (skill.meta.yml), and the eval cases all derive from this file. If a claim is not here, it does not belong in the skill. Drafted by the think-research-framework engine (2026-06-08) and admitted as a Build.

Skill	thinking-framework-skills.contradiction-resolution (installable name think-contradiction-resolution)
Family	problem-framing
Evidence tier	P governing (honest read M/P, capped at P - see “What the evidence shows”)
Confidence	Moderate that the reframe-and-separate discipline surfaces real dissolutions; low that any specific effect size transfers to agents
Status	draft (admitted 2026-06-09 from the SP6 discovery shortlist)

---

1. The mechanism (what actually does the work)

Most problems arrive as a trade-off: to get more of A you must accept less of B. The default response is to optimize - find the least-bad balance point on the curve. Contradiction resolution refuses that move first. It reframes the problem as a contradiction to be dissolved - a solution in which you stop having to trade A against B at all - rather than a trade-off to be balanced. This reframe is the durable cognitive move, and it is the de-branded core of TRIZ (the Theory of Inventive Problem Solving).

The move has two named parts:

1. State the contradiction precisely. A technical contradiction: improving one parameter degrades another (“thicker armor protects better but makes the vehicle too heavy”). A physical contradiction: a single parameter must hold two opposite values at once (“the landing gear must be present, to land, and absent, to reduce drag”; “the coffee must be hot, to enjoy, and not hot, to not scald”). Naming which form you face is what unlocks the resolution menu, because the two forms resolve differently.
2. Aim at the Ideal Final Result (IFR), then try to dissolve. Before reaching for any mechanism, state the Ideal Final Result: an implementation-free description of the end-state in which the benefit is delivered with no new cost, no harm, and ideally no new mechanism - “the function happens by itself.” The IFR is not a goal restatement; it strips out every “how” so it can expose how much of the assumed apparatus is actually necessary, countering the psychological inertia that makes the trade-off feel inevitable. Then attempt to dissolve the named contradiction with a fixed menu: for a physical contradiction, the four separation principles - separate the opposing requirements in time, in space, by scale / system level, or by condition; for a technical contradiction, the 40 inventive principles (segmentation, asymmetry, nesting, “the other way round”, and so on), classically looked up in a 39x39 contradiction matrix.

The output is a contradiction-resolution worksheet: the trade-off written as an explicit opposing pair, the IFR stated implementation-free, one or more dissolution attempts via the separation/inventive principles, and - critically - an honest exit. If no operator dissolves the contradiction, the worksheet says so and routes the problem onward as a genuine trade-off. The point is not to pretend every trade-off is fake; it is to test, deliberately and with a fixed toolkit, whether this one is, before settling for the balance point.

2. Lineage

• TRIZ (Theory of Inventive Problem Solving): Genrich Altshuller (1926-1998), from 1946, developed from a study of large patent collections. The contradiction concept, the Ideal Final Result, and the patterns of evolution are the three pillars; the 40 inventive principles, the 39x39 contradiction matrix, the separation principles for physical contradictions, and the fuller ARIZ algorithm are the apparatus built on them.
• Naming and IP: “TRIZ” is not a live trademark - the mark was formally abolished as of 12 December 2004 (per the G.S. Altshuller Foundation), and TRIZ is treated as a generic, public-domain term. This skill credits Altshuller as lineage but is not branded and needs no trademark string. It ships under a mechanism-over-brand name, contradiction-resolution, per the library’s first commitment.

3. What the evidence shows, and what it does NOT show

The honest read is split, M/P, and capped at the conservative governing grade of P. Both the split and the cap matter.

What the record supports (the M-leaning half). Unlike many practitioner ideation methods, structured contradiction-based ideation has been put through controlled comparison. Several quasi-experimental studies with student subjects report that TRIZ-style ideation improves the novelty and variety of generated concepts relative to unstructured ideation, while reducing the raw quantity of ideas (the ASEE PEER assessment and related ideation-effectiveness studies). There is also one study close to this library’s actual use context: a human-agent design-collaboration experiment (Cambridge, AI EDAM, n=32) paired designers with an LLM-based TRIZ agent and an LLM brainstorming agent; TRIZ produced higher elaboration and, in the human-agent condition, stronger flexibility (multiperspective thinking).

What the record does NOT support, and why the grade is capped at P. Three honest deductions hold the governing grade down:

1. The evidence is for the broad method, not the specific move under test. The studies measure “doing TRIZ” against “not doing TRIZ.” None isolates “reframe as a contradiction to dissolve + state an IFR” as the active ingredient versus the rest of the apparatus, or versus a plain instruction to look for a win-win.
2. The canonical apparatus is empirically contested. Spreafico and Russo (2018) - a critique from inside the TRIZ research community - report that only roughly 10-15% of problems can be adequately described by the matrix’s standard parameters, and recommend working around the single-lookup use. The patent-corpus pedigree is a derivation story, not a validation of recommendation quality, and must not launder the grade upward. The durable, defensible part is the reframe-and-separate discipline; the 39x39 matrix is a contested heuristic.
3. The one AI-context study is mixed-to-negative on the headline novelty claim. In the AI EDAM experiment, brainstorming scored higher on fluency (3.44 vs 2.34 ideas) and was uniquely advantaged on originality; TRIZ’s wins were on elaboration and flexibility. So the most AI-relevant evidence supports “contradiction methods yield more developed, more multiperspective output,” not “more novel solutions” - a narrower claim than the popular framing.

Net grade: P (governing), honest read M/P. Claim “tests whether a trade-off is real and provides a disciplined menu that often dissolves it”; do not claim a measured improvement in solution quality or novelty for agents.

4. Transferred-evidence flag (required honesty for this library)

Almost all of this evidence is human-subject design-education research; the single human-agent study (n=32, one university, neural and creativity measures only, feasibility of ideas not assessed) is suggestive, not decisive, and its authors say so. The evidence is transferred from human practice and only lightly touched by AI-context testing; it is not validated for autonomous-agent use. Treat the AI value as: the agent makes the classify-and-separate pass cheap and disciplined, resists the optimize-the-compromise reflex, holds the IFR implementation-free, and enforces the honest “this trade-off is real” exit - benefits that do not depend on any unproven outcome claim.

5. When it works / when it fails (drives the eval negative cases and “When NOT to Use”)

Works best when:

• A problem is stated as, or has collapsed into, an “A vs B” trade-off whose necessity has not been tested.
• The opposing requirements plausibly hold at different times, places, scales, or conditions (common in design of physical, process, scheduling, and product-mechanism problems).
• The obvious answer is “pick a point on the curve” and a dissolution would be worth far more than the compromise.

Fails or misleads when (poor-fit / anti-patterns):

• The trade-off is genuinely fundamental (physical law, conservation, regulation, hard external limit). Forcing the move manufactures clever non-solutions; the honest output is “real trade-off” routed to think-decision-option-review. This is the central wall, and the worksheet must enforce the exit.
• You need idea volume, not a frame. The method is convergent on one tension and produces fewer ideas; use think-brainwriting or think-far-analogy-ideation for breadth.
• The tension must be managed, not eliminated (an unresolvable polarity like centralize vs decentralize). That is tension/polarity mapping - surface and navigate the standing balance - not this skill, which treats the trade-off as a defect to dissolve and only declares a polarity when dissolution fails.
• No real opposition exists. If the problem is vague or under-specified rather than two-sided, this tool will invent a contradiction. Frame it first with think-problem-restatement.
• The matrix is treated as authoritative. The 39x39 contradiction matrix is a heuristic idea-prompt, not a correct-answer oracle.

6. Output artifact

The skill must emit a contradiction-resolution worksheet, not prose: the trade-off written as an explicit opposing pair; the contradiction classified technical or physical; the Ideal Final Result stated implementation-free; each dissolution attempt with the operator it used (separation in time/space/scale/condition, or a named inventive principle) and whether it yielded a candidate; and the outcome - a concrete resolution, or an honest “this is a real trade-off” with the onward route. A short summary sits above the worksheet.

7. Sources

1. Genrich Altshuller, The Innovation Algorithm: TRIZ, Systematic Innovation and Technical Creativity (English translation, Technical Innovation Center, 1999), and And Suddenly the Inventor Appeared (TIC, 1996). The foundational statement of contradictions, the Ideal Final Result, and the inventive principles. Foundational / practitioner.
2. C. Spreafico and D. Russo (2018), “Altshuller’s Contradiction Matrix. A Critical View and Best-Practice Recommendations.” Critique from inside the TRIZ research community: reports the matrix describes only ~10-15% of problems via its standard parameters and recommends working around the single-lookup use. The key adversarial source on the matrix’s empirical limits. (Critical literature.)
3. ASEE PEER, “Experimental Assessment of TRIZ Effectiveness in Idea Generation” (conference paper). Controlled student-subject comparison; reports TRIZ improves novelty and variety while reducing quantity of generated ideas relative to a control. The nameable controlled-effectiveness evidence, on human subjects. (P, experimental.)
4. “Comparing TRIZ and brainstorming in human-agent design collaboration: effects on cognitive processes and performance,” AI EDAM (Cambridge University Press), n=32. The closest study to this library’s AI-use context: TRIZ higher on elaboration and (human-agent) flexibility; brainstorming higher on fluency and originality. (M/P, single small study, AI-context.)
5. “Ideal Final Result,” Mycoted (TRIZ reference), and “TRIZ,” Wikipedia. Reference descriptions of the IFR, ideality, technical vs physical contradictions, and the separation principles. (Reference.)

Verification status: The mechanism descriptions (Altshuller, Mycoted, Wikipedia) are well-attested and mutually consistent, as is the trademark-abolished fact (G.S. Altshuller Foundation). The Spreafico and Russo critique (2) was read via summary, not the primary full text (the ~10-15% coverage figure is reported as their critical-view claim, not an independently audited constant); the ASEE PEER effect sizes (3) come from the abstract, and the AI EDAM numbers (4) from a fetched summary. None of these gaps changes the conservative governing grade of P. Any unsourced “TRIZ solves N% of problems” figure is excluded on the evidence rule and does not influence the tier.