What is Innovation Design (ID)?
Innovation Design is a systematic approach to two decisions most organizations make informally: what to innovate on, and how that innovation should behave once it’s out in a real, changing world. It replaces ad-hoc brainstorming with a repeatable six-stage process, so innovation stops being an accident and starts being something you can show your work on.
How is I&D different from R&D?
R&D answers two questions: can we discover it, and can we build it. Those are technical questions about feasibility. I&D answers two different questions: should this exist, and how will it keep behaving correctly as its context changes. Those are judgment questions about value and adaptability — and they sit a level above where R&D operates. R&D tells you a thermostat *can* sense occupancy. I&D decides *whether* it should, *for whom*, and *what it does* when the sensor gets it wrong.
Does that mean R&D is becoming obsolete?
No — R&D remains necessary, but it’s no longer where the differentiation happens. Feasibility work is turning into a ommodity: faster, cheaper, and increasingly automated. What doesn’t automate is deciding which of the many things you *could* build are actually worth building, and designing them to hold up once real people and a moving environment interact with them. R&D will forever remain a part of business.
If R&D and I&D are both AI-assisted, what actually differentiates one company from another?
Innovation itself — specifically, the quality of the choices about what to build and why. When every competitor can prototype at similar speed, speed stops being the differentiator. What separates organizations is whether they’re building the *right* things: correctly scoped to a real user need, designed to keep adapting rather than freezing at launch, and backed by a process that can show, stage by stage, why an idea survived. Execution is converging across the industry. Judgment isn’t. That’s why Innovation — not Research, not Development — is becoming the differentiator.
What is the SAI framework, and where does it fit in?
SAI (Systematic Adaptive Innovation – designed by Makarand Pandit) is the set of five principles used in Phase 2 to check whether a design will keep working after launch, not just at launch: Contextual Adaptation (does it respond to a changing environment), External Element Integration (does it depend on signals it doesn’t generate, and handle their loss gracefully), Inversion (does it flip an assumption to unlock new value), Minimalism (does it strip to what’s essential), and Clustering (do its parts act as one coordinated system). It’s paired with SIT — five structural principles applied in Phase 2 alongside SAI — so a design gets checked both for its configuration and its adaptability.
Is this a one-time workshop, or something you can run again?
It’s built to be run again. Every stage is repeatable: change the feature map and Stages 2–6 regenerate around it; adjust scoring weights and the priority ranking updates instantly; emphasize SAI over SIT for a given round without redoing the whole analysis. Markets, budgets, and priorities shift — a process that has to restart from zero every time one of those shifts isn’t rigor, it’s fragility with extra paperwork.
What kind of organization or product is this for?
Any organization that ships a product or system into a context it doesn’t fully control — physical or digital, hardware or software. It’s most valuable where teams already have strong engineering execution but keep discovering, after launch, that they built the wrong thing well, or built the right thing in a way that couldn’t adapt once real users and a real environment got involved.
Does SAI replace TRIZ or SIT?
No — and it isn’t meant to. TRIZ and SIT are, and will remain, extremely effective for design-time decisions: given a contradiction or constraint, what configuration solves it. SAI doesn’t contradict either one; it adds a layer on top. It sits one step ahead in the process, governing how and when those design-time moves stay responsive once the product is actually out in the world. Used together, static inventive thinking becomes living, adaptive design. SAI actually complements TRIZ as well as SIT.
Should TRIZ, SIT, and SAI be used together, or is one enough?
They’re most effective used step by step, not in isolation. TRIZ supplies the broadest set of structural moves; SIT distills that into five faster, more practical principles for the same design-time decisions; SAI then takes whatever configuration TRIZ or SIT produced and asks how it should keep adapting once it’s live. Each principle works on its own, but combining several —
within SAI, and across all three frameworks — multiplies the possibilities far more than any single framework applied alone. However, it should be a worthwhile exercise to use them alone and study the possibilities. They are all equally exciting.
What does "AI-compatible" mean here, and why couldn't TRIZ or SIT get a product there on their own?
It means the product has a supervisory or input layer built in — something that can take in a live signal, defer to it, or be governed by it — rather than running only on logic fixed at design time. Without that layer, a product can be brilliantly engineered by TRIZ or SIT standards and still be structurally unable to work alongside AI systems that expect to feed it context, corrections, or new objectives in real time. SAI’s Contextual Adaptation and External Element Integration principles exist specifically to design that layer in. Note that SAI complements TRIZ and SIT. It does not contradict them in any way.
SAI allows for external input, and that input can be AI itself. So systems designed with SAI may readily adapt to AI and hence can be a part of AI Workflows.
What is the difference between Innovation Design (ID) and Innovation & Design?
Innovation Design is a systematic innovation approach towards innovation. Innovation & Design
(I&D) is a function like R&D. This may be a bit confusing. Let’s simply say that Innovation Design
(the & is missing here) includes Innovation & Design. It also includes R&D.
Why does this distinction matter more now than it did ten years ago?
Because AI has changed which half of the equation is scarce. Research and development — literature review, simulation, prototyping, even code — is now heavily AI-assisted and getting faster every quarter. The “can we build it not” question is being answered for you, whether your organisation has caught up to that or not. The "should we, and how will it adapt" question is not something AI resolves on its own — it requires judgment about context, consequence, and real users. That question is where the actual competitive advantage now lives.
How does I&D leverage AI more effectively than traditional R&D does?
AI is exceptional at optimizing toward a defined goal — running simulations, generating variants, compressing months of prototyping into days. That’s squarely inside R&D’s job, which is exactly why AI erodes R&D’s advantage: everyone gets access to the same acceleration.
I&D leverages AI differently — using it to explore many possible directions cheaply (feature mapping, structural and adaptive possibility generation, scoring, prioritization) so that human judgment gets applied at the *decision* point instead of being consumed by the *production* point. AI increases the number of options you can consider; I&D is the discipline that decides which one is right and keeps that decision honest as things change.
What does an actual Innovation Design engagement look like?
Three phases. Phase 1, Innovation Strategy, decides *what* to innovate on — the business model, positioning, or problem framing — and ends with a clear innovation brief, not a vague direction. Phase 2, Innovation Design, uses the Systematic Adaptive Innovation (SAI) framework to design *how* the chosen idea behaves in the real world, scored and prioritized on impact, cost, and practicality. Phase 3, Implementation & Scaling, hands the brief to execution teams with checkpoints at prototype, pilot, and post-launch to make sure the design intent survives contact with engineering.
How is this different from a traditional stage-gate process?
Stage-gate assumes you’ll get it right the first time through — when a project is killed at a late gate, the work behind it is usually sunk, and reviving it means starting over. Innovation Design’s six stages are each a function of the one before, so changing an input only re-runs what depends on it. A shifted priority means re-scoring, not a new review cycle. A broken market assumption is caught the moment you update the feature map, not at the next scheduled checkpoint. Less work is thrown away when reality changes.
How do you keep the scoring from being just subjective opinion?
Every possibility is rated explicitly on Practicality, Cost, and Impact, on a visible 1-to-5 scale, tied to which principle produced it and which feature category it belongs to. That doesn’t remove judgment — it makes the judgment traceable. If a domain expert disagrees with a score, that disagreement is captured on the record instead of quietly overridden, and for safety-critical or regulated work, the scores are treated as a structured starting point for expert review, not a substitute for it.
How did TRIZ, SIT, and SAI come about?
TRIZ, developed by Genrich Altshuller and first published in 1956 (though it didn’t become widely popular until the 1990s), gave designers 40 inventive principles for resolving technical contradictions. In the mid-1990s, Dr. Roni Horowitz and Dr. Jacob Goldenberg distilled that into SIT (Systematic Inventive Thinking) — just five principles that made the same underlying logic far more practical to apply. SAI (Systematic Adaptive Innovation – designed by Makarand Pandit)) is the newest of the three. It is built specifically to address something neither TRIZ nor SIT was designed to handle: how a product keeps working after it ships, inside a context that keeps changing.
Why weren't TRIZ and SIT enough on their own?
Because neither was built to answer “how does this stay relevant as its context changes in real time?&” Both are excellent at generating a configuration that resolves a known contradiction at the moment of design. Neither one prompts a designer to think about role or design adaptation after that point, and neither accounts for an external input or a supervising layer — exactly what a modern, AI-compatible system needs. That gap is precisely what SAI was built to close. Note that TRIZ and SIT are both more than 50 years old. They tried to cope with challenges of that era. SAI responds to the AI-enabled era of industrial evolution. This is the era of
interconnected systems that will adapt and continue to evolve. SAI as an Innovation Framework supports evolution. In fact, you can find its remote connection to Charles Darwin’s Theory of Evolution. Where a species evolves in response to changing conditions in the environment (that it cannot control). In fact, the Theory of Evolution says that the species that evolves will survive.
Why does taking external input matter so much for modern products?
Because most products no longer operate as standalone tools. Early industrial design produced objects that did one job in isolation. The next stage added feedback loops — automated factories, control systems, complete solutions. We’re now entering a third stage: products designed as nodes inside larger, constantly shifting systems, where the real question isn’t “what does this product do?” but “what else can it do?” A product that can’t take input from outside itself — a sensor, a peer device, a service it doesn’t control — can’t function as a node in that kind of system. Neither TRIZ nor SIT had to account for this, because purely design-time
thinking never required it.
SAI responds to the AI-enabled era of industrial evolution. This is the era of interconnected systems that will adapt and continue to evolve. SAI as an Innovation Framework supports
evolution. In fact, you can find its remote connection to Charles Darwin’s Theory of Evolution. Where a species evolves in response to changing conditions it cannot control. The difference being, the Theory of Evolution is one-way. Whereas SAI can create multiple adaptation scenarios. There can be multiple ways of adapting from one to another. Even reverting to the original scenario is supported.
What about Design Thinking (DT)? How does it compare with Innovation Design (ID)?
Design thinking focuses on empathising with the user. That’s the first step. “Innovate” comes as the third step in the process. In fact, it comes after the “define” step. This has an inherent
problem. This has an inherent problem. How can you define something without a clear innovation strategy? Innovation has a divergence (brainstorming possibilities) followed by a convergence (focus on ideas).
Those who undertake design thinking without a clear innovation strategy often end up with a low success rate or too many options to work on, and may get lost in the process. It should be mentioned here that design thinking is not a one-way pipeline. It loops back. So some of these points may be considered in the next loop.
Innovation should naturally lead to design thinking. The opposite approach may be tricky. We are not saying the process is wrong. We are saying the approach is different. And hence the outcome may be different.