26 General-Purpose Problem-Solving Techniques Applicable Across Domains

**The Case for Universal Problem-Solving**

Most people treat problems as fundamentally domain-specific — a business challenge requires business instincts, a math proof requires mathematical training, a personal crisis requires emotional intelligence. Spencer Greenberg, founder of Clearer Thinking, challenges that assumption directly. His argument is that a set of roughly two dozen general-purpose cognitive strategies can be deployed across virtually any difficult problem, from proving theorems to navigating career decisions to building startups. The techniques are not vague motivational advice but specific procedural moves, each with a distinct logic and concrete examples drawn from mathematics, entrepreneurship, psychology, and everyday life.

What makes the framework compelling is how it reframes the experience of being stuck. Rather than treating a mental block as evidence that a problem is simply too hard, Greenberg treats it as a signal that the wrong strategy is being applied. Switching techniques — not grinding harder on the same approach — is often the key move.

**The Core Toolkit: Restructuring the Problem Itself**

Several of the most powerful techniques operate not by attacking a problem directly but by changing what the problem actually is before any solving begins.

Clarifying is the first and most underrated move. Many people discover, when forced to write out precisely what they are trying to solve and what constraints a valid solution must satisfy, that their mental model of the problem was wrong. A person convinced they need a new job may realize, upon careful articulation, that the real issue is a single fixable relationship with a manager. A startup convinced it has a user acquisition problem may uncover, through rigorous definition, that users do want to sign up — they just abandon the product because of poor interface design. The act of precise description reorganizes information in ways that intuition alone cannot.

Reframing inverts the problem's perspective. One vivid example: a person uncertain whether to take a $20,000 raise to relocate to New York should ask not "should I move?" but "if I already lived in New York, would I take a $20,000 pay cut to leave?" The asymmetry in how those two questions feel reveals that anchoring to the current default — not the underlying merits — is driving hesitation. In a startup context, the same technique appears as a clean-slate question: if the company were being built from scratch today, what would be done differently, and can the product pivot in that direction now?

Unconstraining temporarily suspends the hardest requirement in a problem and solves the easier version first. A company needing someone with expertise at the intersection of machine learning and cryptography might drop the cryptography requirement, recruit deeply skilled machine learning candidates, and then select the one most capable of learning the adjacent field. A software engineer implementing a difficult algorithm might build an inefficient version first and optimize later. In both cases, the unconstrained solution becomes a scaffold toward the real one.

Reexamining takes a more systematic approach to the same territory: listing every assumption embedded in the problem and interrogating each one. A psychology lab that assumes anger always has a cognitively identifiable trigger might miss the role of physiological states like hunger. A mathematician who assumes a function must be smooth might realize that relaxing that assumption immediately makes construction possible using simple glued-together linear pieces.

**Leveraging Other Minds and Existing Knowledge**

A second cluster of techniques is built on a simple insight: most hard problems are not original. Someone, somewhere, has likely encountered something structurally similar and either solved it or written about the attempt.

Searching treats Google Scholar, Q&A sites, and the broader internet as a first-resort oracle. An advertiser struggling to identify the right keywords to bid on can look for blog posts describing keyword strategy for adjacent products. A machine learning practitioner who cannot get a neural network to train correctly can search for tutorials applied to related problem types. The implicit point is that researchers, bloggers, and practitioners collectively constitute an enormous externalized problem-solving resource that many people underuse.

Reading takes a more structured version of the same approach. Rather than improvising a technique for estimating demand elasticity, an entrepreneur can simply open an econometrics textbook — economists have almost certainly solved this. Rather than trial-and-erroring through depression, a struggling person can read well-regarded clinical literature. The technique sounds obvious, but the tendency to reinvent wheels is widespread.

Crowd-sourcing and Experting and Eggheading form a spectrum of social problem-solving. Crowd-sourcing broadcasts a problem widely — a Facebook post, a Quora question, emails to ten relevant contacts — to gather diverse heuristics from people with lived experience. Experting targets someone with domain-specific credentialing. Eggheading goes to the smartest person known to the problem-solver, regardless of domain expertise, and works best when the problem framing is sent in advance so the conversation can begin at depth rather than with exposition.

**Iteration, Intuition, and the Subconscious**

The third major cluster embraces movement over perfection. Guessing — deliberately starting with an approximate answer and stress-testing it — is described not as a fallback but as a legitimate strategy. A startup that does not know what price to charge for its product can pick a number, begin selling, and adjust based on whether price sensitivity shows up in conversations. A mathematician might guess that a differential equation takes the form x^a multiplied by e^(bx), plug it in, and see whether it holds for any values of the free parameters.

Experimenting is the startup-world elaboration of the same idea: build cheap, fast prototypes, run them past five plausible users, and iterate aggressively based on responses rather than debating features in the abstract.

Distracting is the most counterintuitive technique in the set — and has the most distinguished historical pedigree. The method involves saturating working memory with everything relevant to a problem, then deliberately stepping away from it to walk, swim, cook, or nap. The mathematician Henri Poincaré documented one of the most famous instances of this: the recognition that the transformations defining Fuchsian functions were identical to those of non-Euclidean geometry arrived not at a desk but at the instant his foot touched the step of an omnibus, after he had stopped thinking about mathematics entirely. The implication is that the subconscious continues processing in ways that conscious effort can actually impede.

Taken together, the 26 techniques represent a meta-skill: the capacity to recognize when a particular problem-solving mode is failing and to deliberately switch to another. The framework's domain-agnosticism is its core claim — and the reason it applies as readily to a startup pivot as to a mathematical proof.