Of the three terms, only one names a single skill. Analytical thinking is a real capability with nameable parts you can practise and assess; critical thinking is a disciplined use of one slice of it, aimed at judging whether something holds up; and problem solving is a sequence that runs through four separate skills rather than a skill in its own right, which is why "good problem solver" on a job ad predicts so little about how someone will actually perform.
That distinction matters more than it sounds, because these three words do an enormous amount of work in hiring, in school reports and in the way people describe themselves. The World Economic Forum's Future of Jobs Report 2025 found that analytical thinking remains the top core skill for employers, with seven out of ten companies considering it essential. On the graduate side, NACE's Job Outlook 2025 survey of US employers found nearly 90 per cent looking for evidence of problem-solving ability on a student's resume, making it the single most sought-after attribute. Two of the most in-demand things in the labour market, and the terms describing them are used interchangeably by almost everyone.
Here is what actually separates them.
Is analytical thinking a skill?
Analytical thinking is a genuine single skill: the ability to take something apart in order to understand it. The material already exists, and the work is seeing inside it accurately. It has nameable components, it improves with structured practice, and someone's strength in it can be observed directly, which is what distinguishes a skill from a label.
In the Common Codes framework, analytical thinking maps closely onto one cluster of capabilities. Breaking information down sits at the centre of it: decomposing a problem, parsing a document, pulling a system into its component parts. Alongside that sits categorising and classifying, and deciphering meaning, which covers interpreting a set of results or reading what a piece of writing implies rather than states. Testing claims against evidence belongs here too, as does identifying relationships, whether that's a causal chain, two variables moving together, or a comparison between alternatives.
The more demanding end covers examining implications, which is asking what genuinely follows from something and what the knock-on effects will be; appraising quality, where a piece of work or a body of evidence gets judged against a standard; deconstructing arguments, which means exposing the assumption nobody stated and isolating the weakest step; and quantitative reasoning.
Competence in analytical thinking looks like precision. A strong analyst can tell you which part of a thing is doing the work and which part is decoration, and can say "this fails here, on this step" rather than "I'm not convinced". Notice that none of this requires producing a replacement, which is the clean boundary between analysing and the creative work of innovating and synthesising.
What is critical thinking, then?
Critical thinking is a stance rather than a separate skill, assembled from the evaluative slice of analytical thinking plus the habit of gathering evidence. It judges. It asks whether a claim survives contact with the evidence, whether an argument's steps hold, and whether the quality of something meets the standard it's being held to. Critical thinking does not have to build anything, and that limitation is the whole point of it.
Four things do most of the work. Testing claims against evidence means checking a statement against what's actually known rather than against what sounds right. Appraising quality means holding work, arguments and evidence against a standard, which requires knowing the standard. Deconstructing arguments means finding the hidden assumption and the weakest step, so that a critique lands on a specific joint rather than on a general feeling. The fourth component comes from a different place: gathering evidence, which belongs to investigating rather than analysing, because you cannot test a claim against evidence you haven't got.
So critical thinking is analytical thinking pointed at a particular question, which is whether this thing is any good. Someone can be a formidable critical thinker and a mediocre analyst overall, if they're excellent at spotting weak arguments but poor at decomposition or quantitative reasoning; the reverse happens too, where a strong analyst dismantles a system beautifully and then accepts a dubious claim because it came from a senior person. The gap between them is usually disposition rather than capability, since being willing to say the emperor has no clothes is a different thing from being able to see it.
The practical consequence for anyone writing a job ad is that "strong critical thinking" almost never means what the writer thinks it means. It reads to candidates as general cleverness, so they all tick the box.
Is problem solving a skill?
Problem solving is a pipeline that runs through four different skills, and the handover between stages is where it usually fails. Someone can be excellent at three stages and fall over at the fourth, and because "problem solving" names the whole relay rather than any single leg of it, the label tells you nothing about which leg the person can run.
The four stages come from four distinct skill sets.
Analysing does the decomposition. What is the actual problem, what are its parts, and which part matters? A surprising number of failed projects are failures at this stage, where a team solved a real problem competently and it just wasn't the problem they had.
Investigating traces causes, diagnoses faults and isolates variables. This is the diagnostic leg, where you find out why the thing is happening rather than what it is. Mechanics, doctors and engineers live here, and so does anyone who has ever worked out which of six changes broke the build.
Deciding weighs options, ranks priorities, accepts trade-offs and commits under uncertainty. This leg is where the analysis stops and something has to be chosen despite incomplete information. It's the most commonly missing stage in people who are otherwise strong thinkers, and it looks like endless further investigation rather than an obvious failure, which is why it goes unnoticed for so long.
Innovating reframes the problem and generates alternatives. When the options on the table are all bad, this leg produces one that wasn't on the table, often by restating the question so a different kind of answer becomes possible.
Consider a shop losing money on a delivery service. Analysing separates the delivery costs from the product margins and identifies which orders lose money. Investigating traces why, and finds that a single postcode with a low average order value absorbs most of the driver hours. Deciding accepts the trade-off of losing those customers, or of raising the minimum order and annoying some of them. Innovating asks whether delivery is the right model for that postcode at all, and proposes a weekly collection point instead. Four different skills, and a person strong in the first two can still stall completely at the third.
This is the reason the NACE finding above is a little sad. Nearly 90 per cent of employers want evidence of problem solving, and the phrase they're using cannot distinguish between a superb diagnostician who never commits to anything and a fast decider who solves the wrong problem confidently.
Critical thinking vs problem solving vs analytical thinking: a comparison
| What it is | What it does | Output | |
|---|---|---|---|
| Analytical thinking | One skill: breaking things down to understand them | Decomposes, classifies, tests claims, appraises quality, reasons quantitatively | An accurate account of how something works |
| Critical thinking | A disciplined use of part of analytical thinking, plus evidence gathering | Judges whether claims, arguments and work hold up | A verdict, with the weak step named |
| Problem solving | A relay through four skills: analysing, investigating, deciding, innovating | Defines the problem, finds the cause, chooses, and generates alternatives | A change made in the world |
The relationships run in one direction. Analytical thinking supplies critical thinking with most of its machinery, and problem solving draws on analytical thinking for only its first leg, which is why strong analysts are not automatically strong problem solvers. It also explains a pattern most people recognise from work: the colleague who can find the flaw in any plan within thirty seconds and has never once produced a plan. That person has excellent critical thinking and an incomplete problem-solving pipeline, and both of those facts are useful to know about them.
Why does the distinction matter at work and at school?
Naming the specific skill changes what you can do about it. A vague label gives you nowhere to go, while a named component tells you what to practise, what to ask about in an interview, and what to write in a report that a parent can act on. This is the same reason calling them 'soft skills' does more harm than good.
For hiring, the fix is to say which leg of the relay the role actually needs. A compliance role leans hard on critical thinking, since the job is appraising work against a standard and finding the step that fails. An operations role that keeps hitting the same recurring fault needs investigating. A role where the analysis is already done by someone else and the problem is that nothing ever gets chosen needs deciding, and hiring another brilliant analyst will make it worse. Interview questions can follow the same split: asking a candidate to describe a time they committed to a decision they weren't sure about tests something completely different from asking them to pull apart a case study. There's more on this in our piece on hiring for capability rather than job titles.
For school reports and self-description, the payoff is bigger. "Good problem solver" tells a parent nothing at all, whereas "takes a brief apart accurately and traces faults well, but finds it hard to commit when the evidence is incomplete" tells them exactly where to help. The same goes for describing yourself: if you have ever found one part of a project effortless and another part strangely, disproportionately hard, the difficult part probably drew on a different one of these four skills, and that's information rather than a character flaw. Our guides on working out what you're actually good at and on proving skills before you have formal experience both start from that principle, and the same pattern of one umbrella word hiding several distinct capabilities shows up in communication, collaboration and relating.
How do you get better at each one?
Each of the three builds differently, which is another argument for not treating them as one thing.
- Analytical thinking is built through structured practice against feedback. Take arguments and systems apart, state where they fail, and have someone who knows check whether you were right. It responds well to explicit method, since most of the skill is knowing what to look for.
- Critical thinking is built by appraising real work against real standards, out loud, where you can be wrong in front of someone. Reading a rubric and applying it to twenty pieces of work develops it faster than any amount of reading about logical fallacies, because the hard part is calibration rather than knowledge.
- Investigating is built by chasing causes to the end rather than stopping at the first plausible explanation. Any diagnostic hobby will do it, and so will the discipline of asking why one more time than feels necessary.
- Deciding is built by making decisions on a deadline with incomplete information and then reviewing them honestly. The only way to get comfortable committing under uncertainty is to commit under uncertainty, repeatedly, and to notice that most decisions are recoverable.
- Innovating is built through volume and cheap failure, in an environment where a bad idea costs nothing. Constraints help far more than blank pages.
Analytical thinking is a skill, critical thinking is a disciplined use of part of it, and problem solving is a relay of four skills where the baton usually gets dropped between the legs. Working out which leg you're on is most of the work, and it's a great deal more useful than deciding whether you're a "good problem solver". If you want the underlying logic for why a capability has to be specific, nameable and buildable to count at all, that's the subject of our piece on what makes something a human skill.