TL;DR: Body language matters, but the statistic everyone quotes is wrong. The claim that 80% of communication is body language comes from an over-generalisation of two studies Albert Mehrabian published in 1967. Mehrabian has explicitly corrected this misuse, stating that the equations were derived from experiments on communications of feelings and attitudes, and that unless a communicator is talking about their feelings or attitudes, the equations do not apply. The accurate reading is more useful anyway: when your words and your body contradict each other, people believe your body. That makes body language a question of consistency rather than decoration. The 13 mistakes below are framed that way, and the CEOtudent Context Weighting Table shows why the same mistake costs more in an interview than in a casual conversation. Understanding the principle beats memorising the list.
Almost every article you have read about body language opens with the same statistic: 80% of communication is nonverbal, or sometimes 93%.
We are not going to open with that number, because it is not true. But knowing where it came from is the best way to understand how body language actually works.
Fixing the myth first
The numbers come from real research. In 1967 the psychologist Albert Mehrabian published two studies: “Decoding of Inconsistent Communications” and “Inference of Attitudes from Nonverbal Communication in Two Channels”. Out of them came the famous split: words 7%, tone of voice 38%, facial expression 55%.
The problem is that the conditions producing that split got forgotten. The studies examined single words delivered with emotional loading, in situations where the verbal message and the nonverbal message contradicted one another.
Mehrabian himself has had to correct the misreading, noting that the equations were derived from experiments dealing with communications of feelings and attitudes, and that unless a communicator is talking about their feelings or attitudes, these equations are not applicable.
So no, nothing in that research implies your content is only 7% of a technical presentation.
What survives is a more useful principle: when there is a contradiction, people believe the body. If your words say “I support this idea” while your arms are folded and your eyes are elsewhere, your audience takes the second signal. Body language matters not because it replaces words, but because it exposes inconsistency.
Read the 13 items below through that lens. Each one is a problem because it risks contradicting what you actually mean.
Context decides what a mistake costs
The second problem with body language advice is that it presents every rule as equally binding everywhere. In practice, the same behaviour carries very different costs in different settings.
| Mistake | Job interview | Presentation | Video call | Casual conversation |
|---|---|---|---|---|
| Avoiding eye contact | High | Medium | High (not looking at the camera) | Low |
| Closed posture (folded arms) | High | Medium | Medium | Low |
| Excessive hand movement | Medium | Low (often helpful) | High (breaks the frame) | Low |
| Restless fidgeting | High | High | High | Low |
| Poor posture, slumped shoulders | Medium | High | Medium | Low |
| Facial expression contradicting words | High | High | High | Medium |
CEOtudent editorial framework. The weighting is an editorial assessment of the contexts where verbal and nonverbal consistency matters most; it is not a measured statistic.
Two practical conclusions follow. First, applying the same checklist everywhere is unnecessary load. Second, the final row is high everywhere. Inconsistency is the most expensive mistake regardless of context, which is precisely what Mehrabian’s research genuinely showed.
The 13 mistakes
1) Excessive hand movement. Gestures support what you say until they compete with it. On video calls the effect is amplified, because the frame is tight.
2) Folded arms. Usually read as defensiveness or being closed off. You may just be cold; the other person does not know that.
3) Avoiding eye contact. Creates an impression of low confidence or disinterest. Online, the equivalent is looking at the camera rather than the screen.
4) Unbroken staring. The opposite extreme is also uncomfortable. Natural contact is made, briefly released, and remade.
5) Restless fidgeting. Spinning a pen, bouncing a leg, touching your hair. These give away tension faster than anything else, and are usually unconscious.
6) Slumped shoulders. Posture carries much of the signal you send about yourself. Standing straight does not mean standing rigid; rolling the shoulders back is enough.
7) Checking your phone or watch. The fastest way to tell someone you would rather be elsewhere.
8) A facial expression that contradicts your words. The most expensive item on the list. If you say something positive while your face says otherwise, your face wins.
9) Misjudging distance. Too close is uncomfortable, too far reads as detached. Norms also vary between cultures.
10) A weak or crushing handshake. Both leave an impression, and neither is the one you want. Firm without force is the target.
11) Turning your torso away while speaking. The direction your body faces is read as the direction of your attention. Even where your feet point gets noticed.
12) Over-nodding for approval. Constant nodding can look like listening, but more often reads as eager agreement or unease. Occasional and genuine is stronger.
13) Rushing to fill silence. Filling every pause is the nonverbal equivalent of low confidence. Short silences usually work in the speaker’s favour.
Lead like a CEO, learn like a student
Leading yourself like a CEO means matching your signal to the context. That is not performance, it is clarity: what you intend to say and what your body transmits should point in the same direction. Good leaders do this so the message is not lost in noise, not to manipulate anyone.
In practice that means not memorising the list but asking one question regularly: does how I look contradict what I mean? That question covers all thirteen rules at once.
The student side is about noticing the myth at the top of this article. The 80% figure was repeated for decades because nobody checked the source. How many people repeat a claim tells you nothing about whether it is true. The habit of going back to the source pays off far beyond body language, and we applied the same source-checking logic to a different topic in morning routines: evidence versus influencer theatre.
One closing caution: reading body language is not mind reading. Drawing firm conclusions from a single behaviour is a mistake. Someone with folded arms may be defensive, or may be cold. Signals mean something only in clusters, and only in context.
Frequently asked questions
Is 80% of communication really body language?
No. That figure comes from an over-generalisation of two 1967 studies by Albert Mehrabian, which examined single-word emotional communication where the verbal and nonverbal messages conflicted. Mehrabian has explicitly stated that unless a communicator is talking about their feelings or attitudes, these equations are not applicable. The accurate principle is that when signals conflict, people believe the body.
Does body language still matter then?
Yes, but for a different reason. It matters not because it replaces your words, but because contradicting them damages your credibility. Consistency is the real issue, not performance.
Do I need to think about all 13 at once?
No, and trying usually backfires. Managing thirteen behaviours simultaneously destroys naturalness. The more workable approach is a single question: do how I look and what I mean point in the same direction?
Do the rules change on video calls?
Some change noticeably. Eye contact means looking at the camera rather than the screen, hand movements are more distracting in a tight frame, and only your upper body is visible. Those differences are shown separately in the context table.
Can I tell what someone is thinking from their body language?
Not reliably. A single cue can have several causes, and confident interpretations in popular sources are usually overstated. Reading signals as a cluster and in context is far more accurate than drawing conclusions from one behaviour.
Sources
- Albert Mehrabian and Morton Wiener, Decoding of Inconsistent Communications, 1967
- Albert Mehrabian and Susan Ferris, Inference of Attitudes from Nonverbal Communication in Two Channels, 1967
- Albert Mehrabian, Silent Messages, Wadsworth
- Albert Mehrabian’s own public clarifications on the limits of applying the equations
- Erving Goffman, The Presentation of Self in Everyday Life, Anchor Books
- American Psychological Association, reference publications on nonverbal communication
This content was compiled with the support of AI following in-depth research, then written and prepared for publication by the CEOtudent editorial team.
This post is also available in:
