Socionics Signal

What's new in the second edition of "Your Social World Explained"

Spencer Stern — Tue, 06 Oct 2026 00:00:00 GMT

"Your Social World Explained" is back in a fully revised second edition. A few of you asked what's actually different, so here's the honest answer.

Why a second edition at all?

The first edition came out years ago and then sat quietly on Amazon. When I reopened it this year, I did what I'd do with any old analysis: I checked it against its own rules.

The diagrams in socionics aren't decoration. They follow from the model, so I regenerated every one of them from the type data rather than redrawing them by eye. That turned up a few things I'm not proud of:

The Benefit ring ran in the wrong direction.
Two of the "square" relationship groups had the wrong types in them.
Some of the dates for the people behind the theory didn't match the sources.

All three are fixed. If you learnt any of those from the first edition, sorry. This is exactly why I check.

What else changed

A restored chapter. A chapter had dropped out of the Kindle edition at some point. It's back.
Every graphic rebuilt. Over 80 diagrams are now consistent, accurate and readable, with the four quadras in fixed colours throughout.
New material. There's a diagram for the Phonetics small group, an expanded conclusion and a longer list of recommended resources, with the details checked.
A hardcover. The book is now available on Kindle, in paperback and, for the first time, in hardcover, with the hardcover in full colour.
A clearer subtitle. It now says what the book is: a complete guide to socionics, the 16 personality types and relationship compatibility.

What's not in it yet

Small groups. There are 35 in total, and the book covers only some of them. All 35 are already on this site, and they'll go into a third edition next year.

Not sure of your type?

There's a free sample chapter for each of the 16 types, covering your type's chapter and the chapter on duality. Start there.

Thanks for reading, and for the reviews so far. They make a real difference to a book like this.

Spencer

Third for "socionics"

Spencer Stern — Wed, 16 Sep 2026 00:00:00 GMT

What page one looks like for a one-year-old site, and what the results above it reveal.

I ran a clean test this week. Incognito window, no account signed in, no personalisation. I typed the single word "socionics" into Google.

Sociotype came first. Wikipedia came second. Socionics Insight came third.

Sociotype has been publishing since the early 2010s. Wikipedia is Wikipedia. Socionics Insight has existed for less than a year.

I want to be careful here, because rankings move and a screenshot is a snapshot rather than a trophy. Position three today can be position seven next month. But it's worth pausing on, because the three results together say something about the state of socionics in English.

Three descriptions of the same subject

Sociotype leads with the encyclopaedic framing: a personality typology from the 1970s that emphasises information metabolism and interpersonal relationships. Accurate, and completely inert if you've never heard any of those words.

Wikipedia leads by calling it a pseudoscientific theory of information processing and personality types. That's the dismissal, and it's the first thing most newcomers read. (I've written about that question properly in Is socionics scientific?)

Socionics Insight leads with a question: what actually happens between two people. That's the reason the site exists. Every other framework I've used profiles the individual and then leaves you to guess at the dynamic. Socionics is the only one that treats the pair as the thing being studied, and that's the part almost nobody explains in plain English.

So the results are, in order: the reference, the dismissal and the practical. Google has effectively laid out the whole conversation in three lines.

There's a second detail in that screenshot that matters more to me than the ranking. The AI Overview underneath cites Socionics Insight alongside Wikipedia for its explanation of the eight functions. When someone asks an AI what socionics is, part of the answer now comes from material written here.

How it got there

Nothing clever. No link buying, no keyword games. Three things, repeated:

Write the pages a beginner actually needs, in the order they need them.
Source every type and relation from one canonical dataset, so the site never contradicts itself across pages, which is the failure of most socionics resources online.
Cut hard. The sitemap went from 372 pages to 206 this summer, because weak pages drag down the good ones.

That's the whole method. It's slow and unglamorous, and it works because the competition for clear explanation is thinner than it looks.

Which is the real point. If a site under a year old can reach page one for the main search term, it isn't because the site is remarkable. It's because the gap is. Most socionics material in English is translated, unattributed, contradicts itself, or is locked inside forum shorthand that assumes five years of prior reading. Demand for a clear explanation is well ahead of supply.

So thank you to everyone reading this, because a good share of that traffic started with you sharing a link in a Discord server or a Reddit thread.

And a question. If you sent a curious friend to the site tomorrow, where would they get stuck? That's the next thing I'll write.

Read the guide: the beginner's guide to socionics

What 20,000 monthly pageviews actually looks like

Spencer Stern — Mon, 17 Aug 2026 00:00:00 GMT

Update, November 2026: display ads went live on the site on 3 November 2026, and traffic has since passed 25,000 pageviews a month.

Socionics Insight is now doing roughly 20,000 pageviews a month. In April it was around 5,000.

That's the headline, and headlines are the least interesting part. What follows is the version I'd have wanted to read a year ago: what actually moved, what I got wrong, and why growth and revenue turn out to be two separate problems.

The growth

Nothing clever happened. No viral post, no backlink campaign, no growth hack.

What happened was consistent publishing on a subject almost nobody covers properly in English. Socionics has a serious literature in Russian and Ukrainian, a scattered presence on a handful of forums, and very little that a curious English speaker can land on and trust. That gap has been the whole strategy.

The search numbers tell the story more precisely than the pageview count. Average position moved from 9.34 to 6.25 over the same period. Click-through rate roughly doubled, from about 1% to 2.17%. Those two things compound: better positions show you to more people, and better titles convert more of them.

The counterintuitive bit: I deleted a third of the site

At the end of July, Socionics Insight went from 372 indexed pages to 206.

I want to be careful about how I describe why, because I first described it wrongly to myself. My first instinct was that the removed pages were "thin content", using low traffic as the measure. That's a bad measure. Content quality and search demand are different things. A page can be excellent and simply answer a question nobody is asking.

The better way to see it is as a ratio: how much of what Google has indexed is actually earning anything. Pages that were crawled and declined, near-duplicate comparison pages, and individual pages that belonged behind a hub rather than in the index. Those went, with redirects where a redirect made sense.

The site got smaller. The traffic didn't.

The part nobody warns you about: geography

Here's where the tidy growth story runs into economics.

Display advertising isn't priced on how many people read you. It's priced on where they are. Most premium ad networks want a majority of your traffic to come from the US, UK, Canada and Australia. Fifty per cent is a common minimum.

Socionics Insight sits at about 27%.

That isn't a failure. It follows directly from what the site is. Socionics has its deepest roots in Eastern Europe and its most active new interest across South and Southeast Asia and Latin America. An English-language socionics resource is going to be read globally. The audience is exactly the one I wanted. It just happens to be one that premium ad networks pay very little for.

So one network was ruled out on those grounds alone. That was a useful thing to learn early rather than after a rejection email.

Where it stands

I'm currently working with Newor Media, a Google Certified Publishing Partner, to get AdSense set up properly. The site-level review is with Google as I write this. Approved isn't the same as pending, and I'm not going to describe it as approved until it is.

What I would say is that this is deliberately a baseline, not a business. Ads on a site with this traffic mix are a modest passive floor. The real revenue lines are the typing service, the books and Socion, the matching app. Advertising just stops the hosting bill being a pure cost.

What I'm watching next

Three things, all unresolved.

Desktop click-through rate is well below mobile on the same queries at the same positions. The most likely explanation is AI Overviews absorbing US desktop clicks before anyone reaches a link. If that's right, it's structural rather than fixable, and the response is to be the place people go after the summary rather than instead of it.

All sixteen type pages share the same click-through problem: 0.68% to 1.22% at positions six and seven. Sixteen pages failing in the same way isn't sixteen problems. It's one template problem, and it's next on the list.

And the articles section is where most of the crawled-but-not-indexed pages live. That's the clearest remaining sign that something in there isn't earning its place.

The thing underneath all of this

The moat is accuracy. Socionics attracts a technically literate, sceptical audience who will catch an overclaim immediately and never quite trust you again. That applies to the theory, and it applies to writing like this too. Which is why this issue says "in review" rather than "approved", and why the pruning explanation is the one I ended up with rather than the one I started with.

Growth is easy to narrate after the fact. Most of it was showing up.

Two AIs, one app, and why they do very different jobs

Spencer Stern — Mon, 06 Jul 2026 00:00:00 GMT

Most people who meet an AI chatbot inside an app assume it's one thing wearing different hats. Inside Socion there are actually two, and keeping them separate was a deliberate design decision rather than an accident of engineering.

One helps you understand socionics. The other helps work out your type. They never do each other's job, and that boundary is the point.

The first bot: an assistant that won't guess

The "Ask the AI" assistant lives throughout the app, ready for the questions that come up as soon as you start taking intertype relations seriously. What actually happens between a Dual pair day to day? Why does my Mirror keep rubbing me the wrong way? What does my Vulnerable function have to do with the arguments I keep having?

The interesting engineering isn't that it answers. Plenty of chatbots answer. It's that it's built specifically not to improvise. The intertype relations matrix, the sixteen function stacks, the Model A positions and the quadra groupings are all supplied to the model as an authoritative reference it must look up rather than reconstruct from memory. If you ask for the relation between two types, it reads the answer off the matrix instead of reasoning its way there and hoping.

Anyone who has watched a general-purpose AI confidently invent the direction of a Benefit relation, or swap someone's Creative and Role functions, will understand why that matters. Socionics is unforgiving of small errors: one flipped function changes everything downstream. Grounding the model in canonical data is the difference between a reference you can trust and a very fluent guess.

It personalises, too. If it knows your type, it answers from where you actually stand, with your Duals, your Conflictors and your blind spots, rather than reciting theory in the abstract. And when it refers to the wider literature, it points to Socionics Insight, so the trail doesn't go cold.

The second bot: a conversation, not a quiz

The typing chat in onboarding is a different animal, and it tackles the hardest problem in the whole field: most people mistype themselves, especially early on.

The usual fix is a questionnaire: tick sixty boxes, receive a verdict. Socion uses a short, warm, adaptive conversation instead. It moves through areas that tend to be genuinely revealing: how you think through problems, how you decide when the information isn't all there, what drains you socially, how criticism lands, and what you care about that others seem indifferent to. Crucially, it can follow up. If an answer is revealing, it digs; if a topic is exhausted, it moves on.

Underneath, this runs as two separate parts. One conducts the interview, one question at a time. The other reads the finished transcript and produces the actual read: a preliminary type, a confidence figure and the near-miss alternatives. Splitting the roles keeps the interviewer focused on listening rather than half-diagnosing as it goes, which is exactly the trap human typists fall into when they decide too early.

Two features are worth calling out. First, the analysis starts from where functions sit in the stack (Leading, Creative, Role, Vulnerable) rather than from dichotomies, which is the method that holds up best under scrutiny. Second, before it commits, it must check the Mirror of whatever type it's leaning towards. Every type has exactly one Mirror, with the same top two functions in swapped order, which makes it the easiest neighbour to confuse. Forcing that comparison every time catches the most common mistyping before it reaches you.

And it's honest about doubt. High confidence when the evidence is clear; a stated alternative when it isn't; and when two candidates are genuinely neck and neck, it says so and asks you which way you lean, rather than inventing certainty it hasn't earned. It's a preliminary read that sits below a full human-reviewed typing, and it tells you that plainly.

Why the wall between them matters

It would have been simpler to build one bot that does everything. It would also have been worse.

An assistant that answers theory questions wants to be expansive and knowledgeable. A typing interviewer wants to be curious and restrained, giving nothing away while it listens. Combine them and you get a bot that lectures when it should be asking, or types you from a stray comment when you only wanted to understand your Suggestive function. Keeping them apart lets each be good at one thing.

That idea runs through the whole product: the theory is only as useful as the accuracy underneath it, so every layer is built to protect that accuracy rather than paper over it.

Both are live in the app now.

Most team problems are between people, not within them

Spencer Stern — Sun, 07 Jun 2026 00:00:00 GMT

Most personality tools used at work profile individuals. They tell you about Sarah. They tell you about Tim. Then they leave you to work out what actually happens when Sarah and Tim have to ship something together.

Socionics starts from the other end. It describes the relationship directly: not "here's Sarah's type" but "here's the named dynamic that plays out whenever Sarah and Tim are in the same room".

That matters because the problems most founders and team leads lose sleep over are between people, not within them.

The maths of a team

Every pair in a team has its own dynamic. A team of 8 has 28 pairs. A team of 12 has 66. Each one is one of the 16 intertype relations, and each relation has a predictable character.

Individual profiles can't show you any of those 28 or 66 relationships. You only see them by looking at the pairs.

Four places it shows up

1. Conflict nobody can diagnose. Two people don't work well together. Communication training hasn't fixed it. A reorganisation hasn't fixed it. Often the cause is structural rather than behavioural, and once you name the dynamic you know whether to bridge it or route work around it. The Conflict tool shows where the tension points sit for any two types.

2. Founder and leadership pairings. Noam Wasserman's research for "The Founder's Dilemmas", a study of 10,000 founders, found that 65% of high-potential startups fail because of conflict between co-founders (CNNMoney). Very few founders look at how they relate as a pair before signing a shareholders' agreement.

3. Low-level friction that leaks quietly. The product manager and engineering lead who always talk past each other. The mentor and mentee who never quite click. None of this shows up in performance reviews. It shows up in slow delivery and people quietly leaving.

4. Hiring into blind spots. A team that is collectively thin on long-range thinking will keep failing in the same way, and often can't see it, because that's exactly what it's missing. Looking at which functions and clubs a team covers tells you what to hire for before you write the job description. There's more on this in socionics clubs and company structure.

Why this is useful rather than just interesting

The difference is that socionics makes specific predictions you can check.

Not "these two have different communication styles", which is true of almost anyone. Instead: "these two are in a Supervision relation, so one will keep spotting the other's flaws and the other will feel constantly watched and corrected, however much goodwill there is on both sides."

You either recognise that or you don't. In my experience, people usually do, and naming it takes a lot of the personal sting out of it. It stops being "Tim is difficult" and becomes "this pairing has a known pattern, so let's set up the work with that in mind".

Try it on your own team

Pick the two people whose working relationship you'd most like to improve. If you know or can estimate their types, look them up in the Conflict tool or the compatibility tool and see whether the description matches what you've watched happen.

If you're a founder or team lead living with friction you can't explain, I'm happy to take an informal look. Write to me at hello@socionicsinsight.com.

Socion: week three

Spencer Stern — Thu, 16 Apr 2026 00:00:00 GMT

Three weeks in. Here's where things stand.

136 sign-ups. 105 members across all 16 types. 187 connections. 3,773 messages sent.

The headline number this week isn't the growth. It's that all 16 types are now represented on the app. That's the first real milestone, because the theory only works if the full map is there. It is.

The organic baseline is holding. The launch spike faded, as expected, but sign-ups haven't dropped to zero. People are still finding Socion through Reddit, Discord and Socionics Insight. A few have come back a second time to tell other people, which is the thing you hope for and can't really engineer.

The average rating is up to 4.7 out of 5. The messaging volume surprised me: 3,773 messages between 105 members is a lot of active conversation, not dormant connections sitting in a list.

On the product side, the last few weeks have been about infrastructure rather than flashy features: push notifications, reply threading, a network visualisation page, anonymous mode, and a verified badge for typed members. The foundations are solid now.

Also this week: a shop

Something separate launched this week: a shop at shop.socionicsinsight.com, starting with sixteen type mugs, one per type, each with a dictionary-definition design in its quadra's colour. It's a way to show your type without it looking like a personality quiz result from 2014.

Update, November 2026: the shop has since relaunched with a new range at the same address.

Week four starts now. The question I'm watching: do the connections made in week one turn into ongoing conversations, or do they go quiet? Early data suggests the former, but it's still early.

If you haven't tried it: socion.app. Free, no algorithm, no ads. You pick the dynamic.

Spencer

I launched a matching app. Here's why it's different.

Spencer Stern — Wed, 25 Mar 2026 00:00:00 GMT

For the past few months I've been building Socion, a matching app built entirely around socionics intertype relations.

It went live this week at socion.app.

The idea

Most matching products are black boxes. A private algorithm optimises for engagement, not compatibility, and you never know why you're being shown someone.

Socionics is different. It maps 16 named relationship dynamics between every possible pair of types: not just compatible or incompatible, but characterised relationships with predictable patterns. Duals complement each other naturally. Mirrors stimulate each other but tend to criticise. Conflict drains both sides regardless of effort or goodwill.

Socion puts that theory directly in users' hands. You pick your type, choose which dynamics you want to match on, and browse profiles whose type produces that relation with yours. The matching logic is the intertype matrix: open source and auditable.

Why it's interesting as a data problem

Socionics makes predictions that can be tested. If the theory is right, Dual pairs should report better connections than Conflict pairs at scale. Mirror pairs should produce more messages. Certain dynamics should go with longer conversations.

As far as I know, no one has tested this at any meaningful scale in the English-speaking world. The app is designed to generate that data, through post-match ratings, conversation length and connection outcomes, and find out whether the theory actually holds.

That's the product analytics angle that interests me most. Not just building a matching product, but building one that can confirm or challenge the theory it's built on.

What's live

Type onboarding (questionnaire, or bring your own type)
A matching feed filtered by intertype relation
Messaging with real-time delivery
Post-match feedback
Open source code at github.com/sstern42/socionics-match

If you know your socionics type and want to try it, it's free at socion.app.

And if you're a researcher interested in personality type and relationship dynamics, I'd welcome a conversation about what this data could produce.

The reference site is live

Spencer Stern — Thu, 12 Mar 2026 00:00:00 GMT

If you've been following this newsletter for a while, you'll know I've been writing about socionics types, functions and relations in short bursts. This edition is about where all of that now lives properly.

socionicsinsight.com is live.

It's a free reference site covering the full classical socionics system: all 16 types, all 8 functions, all 16 intertype relations, and the small group structures, including quadras, clubs and temperaments. It's built to be accurate, readable and genuinely useful, rather than another surface-level type summary site.

What's there

Full profiles for all 16 types, written through the SLIDE System framework
Every intertype relation explained in plain language
A compatibility tool: pick any two types and get the relation instantly
A type quiz
A dual finder
A growing articles section covering the hidden agenda, socionics subtypes, duality and more

Why it exists

I've been researching socionics since the mid-2000s. The original material sat in an archive for nearly two decades. This is its proper home: maintained, expanded and built to last.

The site presents classical socionics through the SLIDE System framework, which includes a reworked Model A diagram that makes three things visible at once: position strength, the direction information flows, and the divide between conscious and subconscious. The result is an hourglass shape that shows why the model works the way it does, rather than just labelling a grid. You can see it on the About page.

Start here: the beginner's guide

If a type, relation or concept you'd like covered isn't there yet, or isn't covered well enough, tell me. The articles section is where I'm focusing next.

Four small groups that tell you more than a type code

Spencer Stern — Tue, 02 Dec 2025 00:00:00 GMT

A type code like LII or INTj is a summary. It's useful shorthand, but on its own it tells you very little about what someone will actually do in a meeting, a friendship or a relationship. The detail lives one level down, in the small groups each type belongs to.

Socionics has 35 small groups in total. Four of them do most of the practical work, and two of those are enough to pin down any type on their own.

Why the code alone gets misread

When people read a type description, they tend to notice the parts that fit and skip the parts that don't. That's how the typing community ends up full of confident self-typings that don't survive a second look.

Small groups are harder to fool yourself with, because each one describes a single, observable pattern. You can ask "does this person argue for outcomes or for principles?" and get a cleaner answer than "am I an LII?"

The four groups worth knowing

Quadra: what someone values. The four quadras, Alpha, Beta, Gamma and Delta, share a set of valued functions. Alpha values ideas, structure, shared mood and comfort. Gamma values pressure, personal loyalty, results and timing. Quadra is the culture a person feels at home in, and the one they quietly push back against.

Club: what kind of work feels natural. Researchers (NT), Pragmatists (ST), Humanitarians (NF) and Socials (SF). Club predicts the kind of problems someone enjoys solving, which is often more useful for career fit than a full type description.

Romance style: the role taken in close relationships. Careful, Infantile, Aggressor and Victim. The names are classical and a bit blunt, but the pattern is real: each style approaches courtship and partnership differently, and pairs within the same quadra tend to complement each other.

Communication style: how someone comes across in conversation. Passionate, Sincere, Firm and Cool. This is often the first thing other people notice, and the first thing they misread. A Cool communicator isn't necessarily unfriendly. It's just their natural register.

Two groups are enough to identify any type

Here's the part most explanations skip. If you know someone's quadra and their communication style, you know their type. Every combination points to exactly one of the 16:

Quadra	Passionate	Sincere	Firm	Cool
Alpha	ESE	SEI	ILE	LII
Beta	EIE	IEI	SLE	LSI
Gamma	SEE	ESI	LIE	ILI
Delta	IEE	EII	LSE	SLI

The reason is structural. Communication style combines two basic splits: extravert or introvert, and logical or ethical. Passionate types are extraverted and ethical, Sincere types introverted and ethical, Firm types extraverted and logical, and Cool types introverted and logical. Each quadra has exactly one type in each of those four positions.

Club and romance style then work as cross-checks. If your read of someone's quadra and communication style points to LII, but they plainly behave like a Pragmatist, one of your two readings is wrong, and it's worth finding out which.

A worked example: LII and ILI

LII and ILI are one of the most commonly confused pairs, and the small groups show why.

Group	LII	ILI
Club	Researcher	Researcher
Communication style	Cool	Cool
Quadra	Alpha	Gamma
Romance style	Infantile	Victim

They match on the two groups that are easiest to see from outside: both are reserved, analytical and drawn to theory. So people who type from first impressions often can't tell them apart.

The difference is in the two groups that take longer to observe. LII belongs to Alpha, which values open-ended ideas and a relaxed, low-pressure atmosphere. ILI belongs to Gamma, which values timing, practical results and loyalty between individuals. Watch what each one criticises, and what they find a waste of time, and the quadra usually shows itself. There's a full comparison in ILI vs LII: how to tell them apart.

Using small groups when typing someone

Instead of asking which description fits, try answering four narrower questions:

Which quadra's values does this person defend without being asked?
What kind of problems do they choose to work on when nobody is directing them?
What role do they take early in a relationship?
How do they sound when a conversation gets tense?

These are things other people can observe, which is one reason an outside view is often more accurate than self-typing. If you'd like that outside view, a structured typing works through exactly these kinds of signals.

The takeaway

The type code is the label on the box. The small groups are what's inside it. Start with quadra and communication style, use club and romance style to check your answer, and you'll mistype people far less often.

What quadra is your team? Reading the values of a group

Spencer Stern — Wed, 05 Nov 2025 00:00:00 GMT

Most of socionics is about individuals and pairs. Quadras are where it scales up to groups. A quadra is a set of four types that value the same four functions, and groups of people tend to drift towards one quadra's values, whatever the types of the people in them.

You've probably felt this without a name for it. Some teams leave you energised after a long day. Others leave you drained after a short one, even when nothing obviously went wrong. Often the difference is whether the group's values match yours.

Alpha: the room full of ideas

Types: ILE, LII, ESE, SEI. Valued functions: possibilities, logical structure, shared mood and physical comfort.

An Alpha group feels open and easy-going. People bounce ideas around, follow tangents for fun, and care more about whether something is interesting than whether it's useful yet. Everyone gets a turn, nobody pulls rank, and there's usually food involved.

What gets rewarded: curiosity, a good explanation, keeping the atmosphere light. What gets frowned on: pushing people around, or bringing in pressure and status games.

Beta: the team with a mission

Types: EIE, IEI, SLE, LSI. Valued functions: decisive force, logical structure, shared emotion and a sense of where things are heading.

A Beta group has a clear hierarchy and a shared story about why it matters. People know who's in charge, rally around a goal, and feel things together, sometimes dramatically. Loyalty to the group is expected.

What gets rewarded: commitment, toughness, standing up for the team. What gets frowned on: sitting on the fence, or treating the mission as just a job.

Gamma: the group that wants results

Types: SEE, ESI, LIE, ILI. Valued functions: decisive force, personal loyalty, practical results and good timing.

A Gamma group is pragmatic and independent. People judge ideas by whether they work and whether the timing is right. Relationships are one to one rather than group-wide, and trust is earned individually. Rules matter less than outcomes.

What gets rewarded: competence, initiative, delivering. What gets frowned on: theorising without a point, or performing team spirit nobody feels.

Delta: the steady, decent workplace

Types: IEE, EII, LSE, SLI. Valued functions: possibilities, personal relationships, useful work and wellbeing.

A Delta group is calm, fair and focused on doing good work at a sustainable pace. People care about each other's growth and comfort, avoid drama, and build trust slowly. It's the small business that treats staff well, or the team that actually means it about work-life balance.

What gets rewarded: reliability, kindness, quiet improvement. What gets frowned on: showing off, or chasing growth at everyone's expense.

Why the wrong quadra wears you out

Every type has four valued functions and four it treats as background noise. In a group that runs on your valued functions, you don't have to translate yourself. In a group that runs on the other four, you spend the day performing a version of yourself that isn't natural, and that costs energy even when everyone is being perfectly nice. There's more on this in valued and unvalued functions.

This is also why the same person can be a star in one team and a poor fit in the next, with no change in ability. An LII who thrives in an Alpha research group may struggle in a Beta sales team, not because they lack skill but because the group keeps rewarding things they don't value.

A few caveats

Most real groups are mixed. A team often has a dominant quadra set by its leader or founder, with pockets of the others. Groups also shift over time, especially when leadership changes, and the same pattern plays out across whole organisations and cultures. That bigger cycle is covered in how ideas move through the socion, quadra by quadra.

And a group's quadra doesn't tell you anyone's type. A Delta workplace can be full of Gamma types who've learnt to work its way. It tells you what the room rewards.

Try it

Think of the group you spend most time in. What gets praised? What gets someone quietly sidelined? Match that against the four descriptions above, then compare it with your own quadra. If they don't match, it may explain more about your energy at the end of the week than anything else.

What each function looks like from the outside

Spencer Stern — Wed, 22 Oct 2025 00:00:00 GMT

We usually talk about the eight functions as if they live entirely inside people's heads: quiet mechanisms that shape how we see, feel and decide. But nobody has ever seen someone's Introverted Intuition. What other people see is behaviour.

In "Your Social World Explained" I gave each function a second name describing what it looks like from the outside. These are the attitude names used in the SLIDE System on this site. They're useful for one practical reason: you can observe them.

The eight attitudes at a glance

Function	Outward attitude	What you notice
Introverted Sensing (Si)	Dynamic Stabilisation	Keeps things comfortable and notices when they drift
Extraverted Sensing (Se)	Tactical Action	Reads who holds power right now and acts on it
Introverted Intuition (Ni)	Reality Distillation	Senses where a situation is heading before it gets there
Extraverted Intuition (Ne)	Creative Thinking	Sees what something could become, not what it is
Introverted Ethics (Fi)	Ethical Harmony	Keeps a quiet map of who people are to each other
Extraverted Ethics (Fe)	Social Interest	Reads the mood of a room and changes it on purpose
Introverted Logic (Ti)	Academic Knowledge	Checks whether an argument holds together
Extraverted Logic (Te)	Practical Results	Checks whether something works and gets it done

Sensing: comfort and force

Dynamic Stabilisation (Si) is the person who adjusts the heating before anyone complains, remembers how everyone takes their tea, and quietly brings a meeting back to a sensible pace. It isn't control. It's regulation: noticing when everyday experience drifts away from what feels settled, and nudging it back.

Tactical Action (Se) is the person who reads the room for who has power and what's actually happening, then moves. They cut through hesitation and push for a decision now rather than after another round of talk. It can look like dominance. Often it's just that action feels more real to them than discussion.

Intuition: where things lead and what they could be

Reality Distillation (Ni) is the person who is rarely surprised. They condense a messy situation into one line of where it must end up, often before anyone else has noticed it's moving. They can seem detached, but they're compressing experience into a forecast.

Creative Thinking (Ne) goes the other way. Where Ni narrows to one trajectory, Ne opens up many. This is the person who sees potential in people and ideas, jumps between possibilities and gets restless once something is settled. It's less about predicting and more about playing with what could be.

Ethics: inner ties and shared mood

Ethical Harmony (Fi) is the person who knows who's fallen out with whom, who can be trusted, and where they stand with everyone. It's not moralising. It's attunement to the private temperature between people, and it often shows as loyalty and keeping a careful distance.

Social Interest (Fe) is its expressive counterpart: the person who lifts or deflates the mood of a whole room. They express emotion openly and use it on purpose, to make an evening fun or a team feel united. It's how feeling becomes atmosphere.

Logic: structure and results

Academic Knowledge (Ti) is the person who stops a discussion to ask whether a claim is consistent with what was said ten minutes ago. They care about how things fit together more than whether anyone will use them. Logic as symmetry rather than utility.

Practical Results (Te) is the person who asks whether it works, how long it will take and what the evidence says. Where Ti asks "is it true?", Te asks "does it work?", and then organises the effort to get it done.

Why the outward view helps

Two things change when you look at functions this way.

First, the old argument about introverted functions being invisible goes away. Introverted functions work like internal calibration and extraverted ones like external engagement, but both show up in behaviour. They just move at different speeds.

Second, it makes people easier to read kindly. Someone strong in Tactical Action isn't necessarily trying to dominate; they may simply think best by doing. Someone strong in Reality Distillation isn't aloof; they're working out where this is going. The map lets you grant other people the benefit of a different tempo.

It's also the most practical way to start typing someone. You can't interview their unconscious, but you can watch which of these eight things they do without being asked. The functions guide goes into each one in depth, and the About page shows how the attitudes sit in the SLIDE System model.

How to read the intertype relations map

Spencer Stern — Sun, 12 Oct 2025 00:00:00 GMT

Some people click with you straight away. Others seem to speak a different language however hard you both try. Socionics doesn't treat that as luck. It names the dynamic between every possible pair of types, and lays them all out on one map: the intertype relations matrix.

At first glance the matrix is a wall of 256 squares. You don't need to read it all at once. You only need your own row.

What the map actually shows

There are 16 types, so every type has 16 positions on its row: one for each type, including itself. Each position is one of the 16 intertype relations.

Most relations work the same in both directions. If you're someone's Dual, they're yours. Two are one-way, with a giving side and a receiving side:

Benefit: one person is the benefactor, the other the beneficiary. The benefactor feels useful; the beneficiary may feel subtly talked down to.
Supervision: one person is the supervisor, the other the supervisee. The supervisor spots flaws easily; the supervisee feels watched and corrected.

Because each of those has two sides, they take up four positions on every row: you're on the receiving end with one type and the giving end with another.

A worked example: the LII row

Here is the full row for LII, taken from the same data the compatibility tool uses.

Relation	Type	In short
Identity	LII	Deep understanding, but neither can offer what the other lacks
Dual	ESE	The most complementary relation; needs met without effort
Activation	SEI	Energising in short bursts, tiring up close for long
Semi-dual	EIE	Feels like a dual at first; key needs stay unmet
Mirage	LSE	Comfortable and pleasant, slightly idealised
Mirror	ILE	Same top two functions, swapped; each corrects the other
Kindred	LSI	Easy rapport with a competitive edge underneath
Business	EII	Respectful, effective working partners
Quasi-identity	ILI	Similar on the surface, different goals underneath
Extinguishment	LIE	Similar at first; over time each dampens the other
Super-ego	ESI	Mutual respect at a distance
Conflict	SEE	The most difficult relation for both sides
Benefit (LII receives)	IEI	IEI is LII's benefactor
Benefit (LII gives)	SLI	LII is SLI's benefactor
Supervision (LII supervised)	SLE	SLE is LII's supervisor
Supervision (LII supervises)	IEE	LII is IEE's supervisor

Every type's row has the same 16 slots. Only the types in them change.

Start with three relations

If the whole row is too much, start with the three that tell you the most.

Dual. The relation the whole system is built around. Each person's strengths cover the other's blind spots without anyone having to try. When it works, it feels like breathing out.

Conflict. Each person's strongest function lands on the other's most vulnerable one. It isn't about bad intentions. Both sides can be decent and still wear each other down at close range. It works best at a distance, on a shared external task.

Supervision. The one most people recognise once it's named. One person notices everything the other gets wrong; the other feels permanently under review. It can teach a lot, but it's rarely relaxed.

How to use your row

Pick three or four people you know well whose types you're fairly sure of. Find where they sit on your row, then ask whether the description matches what you've lived.

Two things can happen. Sometimes the map names a pattern you'd felt but couldn't explain. Sometimes it doesn't fit at all, and that's useful too: a relation that clearly doesn't match is often the first sign that one of the two types is wrong.

That's the real value of the map. It isn't about putting people in boxes. It's about seeing the currents between them, and why some relationships feel like home while others feel like homework.

To find your own row, open the interactive matrix or your type's relationships page, such as LII relationships.