TRIZ: The Theory of Inventive Problem Solving

TRIZ — pronounced "treez," a Russian acronym for Teoriya Resheniya Izobretatelskikh Zadatch, or "theory of inventive problem solving" — is a creativity method with an unusual origin story. It wasn't developed by a psychologist studying how creative people think. It was built by a patent examiner who read tens of thousands of patents looking for the repeating structure underneath breakthrough inventions.

That examiner, Genrich Altshuller, started in 1946 as a patent inspector for the Soviet navy. His job put him in contact with thousands of inventions per year, and he began noticing something: the same handful of solution patterns kept reappearing across completely unrelated fields. An aerospace patent and a plumbing patent would solve structurally identical problems using the same underlying move. Altshuller's core claim, radical at the time, was that inventive genius wasn't a personality trait. It was a repertoire of transferable moves that could be extracted, catalogued, and taught.

Why TRIZ Exists: The Contradiction Problem

Most engineering problems get solved through trade-offs. You want a product lighter, so you use thinner material, but thinner material is weaker — so you accept some weight to keep the strength you need. That's a compromise, and most design work stops there.

Altshuller's insight was that the best inventive solutions don't compromise. They eliminate the trade-off entirely. A genuinely inventive solution makes the product both lighter and stronger, not lighter or stronger. He called the underlying tension a technical contradiction — improving one parameter (weight) degrades another (strength) — and argued that breakthrough patents were, almost without exception, cases where someone had found a way to resolve the contradiction rather than balance it.

This reframes the goal of problem solving. Instead of asking "what's an acceptable compromise here," TRIZ asks "what would it take to get both, with neither degraded?" That question is uncomfortable, because it initially looks impossible. It's also exactly the question that separates incremental design improvements from genuinely novel solutions — the same territory problem reframing operates in, applied specifically to technical trade-offs.

Built From Patents, Not Theory

What makes TRIZ distinct from other creativity frameworks is its evidence base. Altshuller and later collaborators analyzed a corpus that eventually grew to roughly 2 million patents, filtering for the subset — around 40,000 in his early studies — that represented genuine inventive leaps rather than incremental improvements. From that analysis, he extracted 40 recurring solution patterns, which he called the 40 Inventive Principles.

He continued this work under extraordinary conditions. In 1950, after Altshuller and a colleague wrote a critical letter to Stalin about the state of Soviet inventive practice, he was arrested and sent to a labor camp near Vorkuta, in the Russian Arctic. He spent four years there before being released after Stalin's death in 1953. According to TRIZ historians, he continued developing and discussing his patent-pattern research with fellow prisoners — engineers and scientists swept up in the same purges — during his imprisonment. The theory that would eventually be adopted by Samsung, Boeing, Procter & Gamble, and General Electric had some of its early development done inside a Gulag.

The 40 Inventive Principles

The principles range from concrete engineering moves to broadly applicable cognitive strategies. A few that transfer well outside pure engineering:

Segmentation. Divide an object into independent parts, or make it easy to disassemble. Container shipping is segmentation applied to cargo: instead of loading a ship's hold item by item, break freight into standardized units that can be handled independently. The principle also shows up in distributed computing — breaking a single large task into independently schedulable units — and in this site's own alternative uses test, which segments an object into its component features to find non-obvious uses for each.

Prior action. Perform a required change before it's needed, so the system is ready when the need arrives. Pre-stressed concrete is a direct example: the concrete is placed under compression before it bears a load, so that when tension arrives, it's counteracting a force that's already been accounted for. Pre-loading a plan, drafting a response before a meeting starts, or pre-committing resources before a bottleneck hits are the same move applied to non-physical systems.

The other way round (inversion). Instead of the standard action, do the opposite. Anti-lock brakes are a clean example: rather than adding more friction to stop wheel lock, the system deliberately allows rapid micro-lock-and-release cycles, using controlled slipping instead of preventing it. This is the same cognitive move behind inversion thinking and the "invert, always invert" heuristic that shows up across multiple creativity methods, including problem reframing.

Nested doll (Matryoshka). Place one object inside another, which is inside a third, and so on. Measuring cups that stack inside each other, retractable pens with internal spring mechanisms, and modular furniture systems all apply this principle — each layer solves a storage or access problem the outer layer alone couldn't.

Local quality. Change a uniform structure or environment so different parts serve different functions, rather than forcing one property to work everywhere. A pencil with a uniform shaft but a different material at the tip (the eraser) is local quality in its simplest form. Ergonomic tool handles that vary in texture and shape along their length apply the same logic.

Physical Contradictions and the Separation Principles

Not every contradiction is between two different parameters. Sometimes a single parameter needs to be two opposite things at once — Altshuller called this a physical contradiction. A bicycle spoke needs to be rigid, to hold the wheel's shape, and flexible, to absorb road shock, but rigidity and flexibility are opposite values of the same property.

TRIZ resolves physical contradictions through four separation strategies:

• Separation in space — different parts of the system have different values. The wheel rim flexes slightly while the hub stays rigid; the wing tip flexes while the wing root stays stiff.
• Separation in time — the system has one value at one moment and the opposite value at another. A folding bicycle is rigid while riding and collapsible while stored.
• Separation on condition — the system responds differently depending on context. Smart materials that stiffen under impact but stay flexible under normal use (used in some protective sports gear) separate the contradiction by triggering condition.
• Separation between the whole and its parts — the system as a whole has one property while its components have the opposite. A chain is flexible as a whole even though each individual link is rigid.

This is a more precise version of what creative constraints describes as constraints forcing deeper search: a physical contradiction is a constraint stated so specifically that "flexible AND rigid" stops sounding like nonsense and starts sounding like a design specification once you know which separation strategy to apply.

Ideal Final Result

Before applying any of the 40 principles, TRIZ practitioners are trained to define the Ideal Final Result (IFR): describe the outcome as if it happened with zero cost, zero added complexity, and no new problems introduced. Not a realistic target — a deliberately unconstrained one.

The IFR is a working tool, not a destination. Altshuller's observation was that most people anchor their thinking to the current system and ask how to improve it incrementally. Starting instead from "what if this problem simply solved itself, with the resources already present in the system" reorients the search. A classic TRIZ example: reduce contamination in an environment without adding a filtration system by asking whether an already-present element of the system (gravity, existing airflow, an existing surface) could be reconfigured to do the filtering itself.

This has the same function as the fantasy analogy step in synectics, where you imagine the ideal state without regard for physical constraints, then work backward to find which constraints were actually load-bearing. TRIZ formalizes the same move as a required first step rather than an optional analogy type.

The Contradiction Matrix

For technical contradictions, Altshuller's team built a contradiction matrix: a grid cross-referencing 39 standard engineering parameters (weight, speed, strength, temperature, reliability, and so on) against each other. Each cell in the matrix lists which of the 40 principles have most frequently resolved that specific pairing across the patent corpus.

If you're trying to make something stronger without making it heavier, you look up "strength" against "weight" in the matrix, and it points you toward the small subset of principles — often Segmentation, Local Quality, or Composite Materials — that have historically resolved that exact tension most often. The matrix doesn't generate the solution. It narrows a search space of 40 possibilities down to the 3 or 4 most statistically productive ones for your specific contradiction, which is a meaningfully different task than staring at 40 principles and guessing.

TRIZ vs. SCAMPER vs. Synectics

All three methods constrain search, but they constrain it differently, and the differences determine which one fits a given problem.

SCAMPER applies seven general transformation prompts to any existing object or process. It's fast, domain-agnostic, and doesn't require identifying a contradiction first — you can run it on a marketing plan as easily as a physical product.

Synectics works through structured analogy, deliberately importing structure from an unrelated domain to disrupt fixation. It's most useful when you don't yet have a clear technical specification of what's wrong — just a stuck feeling.

TRIZ is the most structured of the three, and the most demanding. It requires you to state your problem as a specific contradiction between two measurable parameters before the method does anything useful. That precision is a cost — most everyday creative problems don't arrive pre-formatted that way — but it's also why TRIZ produces unusually concrete, checkable outputs in engineering and product design contexts where SCAMPER and synectics tend to produce looser, more exploratory ones.

Applying TRIZ Without an Engineering Background

You don't need the full contradiction matrix to get value from TRIZ's underlying logic. A simplified version:

1. State the contradiction explicitly. Name the two things in tension. Not "how do we make onboarding better" — instead, "faster onboarding reduces the depth of feature discovery." Vague problems don't have contradictions you can work with. Specific ones do.

2. Write the Ideal Final Result. Describe the outcome with the contradiction simply gone — as if the system solved it using resources you already have, at no added cost. Don't worry yet about whether it's realistic.

3. Check whether it's a physical or technical contradiction. If a single property needs two opposite values, look for a separation strategy — in space, time, condition, or whole-versus-parts. If two different properties are trading off against each other, that's a technical contradiction; scan the 40 principles for ones involving segmentation, inversion, prior action, or local quality first, since these transfer best outside pure engineering.

4. Force at least three principles onto the problem, even ones that seem irrelevant. The value of a fixed principle list, the same as with SCAMPER, is that it makes you interrogate angles you wouldn't generate unprompted. Resistance to a principle is often diagnostic — it usually means that principle is pointing at an assumption you haven't examined.

Every TRIZ principle is ultimately a transferred structure — a solution pattern imported from a domain where it already worked. That's the same underlying skill the Analogical Encoding exercise trains directly: recognizing the shared structure between two surface-different scenarios. The more practice you have spotting structural similarity in general, the faster you'll see which of the 40 principles actually fits your specific contradiction.

Where TRIZ Is Used Today

Samsung's adoption of TRIZ in the early 2000s is the most cited corporate case: the company trained thousands of engineers in the method and credited it with a measurable jump in patent output and R&D efficiency during a period of aggressive product expansion. Procter & Gamble, Boeing, LG, and Intel have documented internal TRIZ programs for similar reasons — the method produces defensible, patentable solutions because it's explicitly built from patent data in the first place.

Outside large engineering organizations, TRIZ's core moves — stating contradictions precisely, separating opposing requirements by time or space or condition, defining an ideal result before compromising — apply to any creative problem solving work where the instinct to average two competing goals is quietly producing a worse outcome than resolving the tension between them would. The method's discipline is its value: it won't let you settle for "somewhat better at both" when "both, fully" was available.

Ready to train your creativity? Try science-backed exercises that measure and improve your creative thinking. Start a Free Exercise