Flirting or Just Friendly? How to Tell the Difference
The line is subtle — but these 9 signals make it readable.
Telling flirting from friendliness is harder than the dating-advice industry usually admits, partly because the two share most of their surface behaviors and partly because the people doing the flirting are often not entirely sure themselves. Warmth, eye contact, laughter at small jokes, a hand briefly on a forearm, a question that lingers a beat too long, all of these can be either, and reading them correctly is less about catching a single decisive cue than about noticing a pattern of small differences over time.
The research that bears on this question is older than it might seem. Monica Decades of observational research on courtship in public settings have documented dozens of small flirtation behaviors, most of them initiated subtly enough that the people they were directed at frequently missed them. Mehrabian's work on implicit communication helps explain why these cues carry meaning despite, or because of, being non-verbal. And the broader attachment and social-psychology literature is consistent on a key point: flirting is fundamentally selective, while friendliness is fundamentally distributed, and selectivity is the variable that carries the signal.
The nine signals below are drawn from that combined picture, and none of them is decisive in isolation. What you are reading for is contrast: does this person behave differently with you than with other people in the same room, and does the difference hold across multiple meetings rather than appearing once. The article does not promise certainty, because flirting is one of the most ambiguous human behaviors and is meant to be, but it does try to make the underlying pattern more legible than the standard advice usually does.
Friendliness is consistent; flirting is selective
Friendliness has a steady, even quality. The friendly person is warm to the bartender, the colleague, the stranger asking for directions, and to you, in roughly the same way. Flirting is the opposite shape. It singles out. Energy that would normally be distributed across a room narrows onto one person, and the asymmetry is the signal. Reading the difference well requires watching how the person treats others, not just how they treat you.
Mark spent an evening watching a colleague he had a small crush on. She laughed at his jokes warmly. Then he watched her laugh at someone else's jokes the same way, ask the new intern the same kind of question she had asked him, and offer the same small touch on the arm when saying goodbye. The behavior he had read as private was her standard operating warmth. Painful but useful information.
Eibl-Eibesfeldt's cross-cultural work on greeting signals (Eibl-Eibesfeldt, 1972) catalogued how baseline social warmth — the eyebrow flash, the open smile — is widely shared and not specifically romantic. Romantic signaling tends to involve a narrowing and intensification of these baseline cues toward one target. Mehrabian (1971), often misquoted as the 93% nonverbal claim, did argue convincingly that affective allocation is read primarily through nonverbal channels.
This week, before deciding what someone's warmth means about you, watch them with three other people. Ask whether you are receiving a specific version of their attention or a particularly visible serving of their general warmth. The answer is almost always there within ten minutes of watching, if you can tolerate looking.
Friendly touch is incidental; flirting touch lingers
Touch in a friendly register is incidental and brief. A hand on the back to slip past, a quick squeeze of the shoulder, contact that resolves quickly because it was functional. Flirting touch is different in duration and intention. It lingers a half-second past necessity. It returns to the same spot a second time. It happens when there was no functional reason for it. The information is in the extra fraction of a second, not the touch itself.
Sara reached across Mark at the bar to grab a napkin and her hand brushed his arm. Friendly. A minute later she touched his forearm to make a point, and a minute after that touched the same forearm again to laugh at something he had said. None of the touches were dramatic. But the repetition, the same spot, the absence of any practical reason — that pattern was doing different work than the first reach for the napkin.
Mehrabian's research on the nonverbal communication of attitudes (Mehrabian, 1971) identified what he called immediacy cues: physical orientation, proximity, and lingering contact as honest signals of liking. Eibl-Eibesfeldt (1972) similarly documented that allowed and returned touch in early courtship is one of the most cross-culturally consistent flirting markers. The specifics differ; the lingering doesn't.
Notice this in your own behavior over the next week. With people you find merely pleasant, does your touch resolve quickly. With anyone you find more interesting, does it linger half a second longer. If you can see the difference in yourself, you can usually see it in others. The hand that doesn't immediately retract is rarely accidental.
Both friendly and flirting people ask questions — but flirting questions go deeper
Both friendly and flirting people ask questions. The friendly ones ask the questions a kind colleague would ask: how was your weekend, how's the new job, how is your mother doing. The flirting ones drift toward questions that wouldn't make sense from a colleague. What kind of person were you at sixteen. What's the thing nobody at work knows about you. The friendly question is interested in your information. The flirting question is interested in your interior.
Mark and a coworker had been chatting at a kitchen counter for ten minutes when she asked, what would you do for a year if money wasn't a factor. He gave a real answer. She asked a follow-up. By the time someone else walked in, they had wandered from work logistics into a kind of conversation that did not match the lighting of the office kitchen. Neither of them had touched. The questions had done it.
Aron et al.'s 36-question study on generating closeness (Aron et al., 1997) is the canonical demonstration: escalating personal questions reliably create the felt sense of intimacy, often within an hour. Reis and Shaver's intimacy process model (Reis & Shaver, 1988) clarifies why — specific responsiveness to personal disclosure is the engine. A flirting questioner is, intuitively, running a small version of the Aron protocol.
This week, listen for the shape of the questions you're being asked, not just the warmth of the tone. If someone's questions are pulling you toward the parts of yourself you don't share at work, notice it. And if you want to know whether you're flirting back, notice the shape of your own questions in return.
Eye contact duration is the most consistent signal
Eye contact is the most reliable signal because it's the hardest to fake casually. Friendly eye contact lands and leaves — long enough to feel acknowledged, short enough to feel comfortable. Flirting eye contact holds a beat longer than the conversation requires, returns more often than it needs to, and frequently includes a small downward glance and a return upward. The whole pattern is hard to perform deliberately, which is part of why it reads as honest.
Mark was telling a story at a group dinner. Three people were nodding politely. Sara, across the table, was looking at him in a way that made him lose his place twice. After the story she looked down at her wine, back up at him, and held his eyes for a second too long. He hadn't been thinking of her romantically all evening. He started to.
Hess (1965), in his classic studies on pupil dilation, found that involuntary pupil widening accompanies sustained attraction-laden looking. Eibl-Eibesfeldt's cross-cultural studies (Eibl-Eibesfeldt, 1972) recorded sustained mutual gaze followed by a small downward break as one of the most universally consistent courtship signals. Mehrabian (1971) placed eye contact among the most powerful immediacy cues, more telling than verbal warmth.
This week, in any conversation where you're unsure, notice the duration of the gaze. Friendly will hold for roughly a beat. Flirting will hold for roughly two, and the eyes will keep finding you across the room when the conversation has moved on. Don't escalate just because you noticed. Just notice. The information is enough for now.
Friendly people include everyone; flirting people find ways to narrow the room
A friendly person works the room. They circulate, include outliers, fold the quiet person into the conversation. Their warmth has a centrifugal quality. A flirting person quietly works against this. They drift toward you when the group reforms. They pick the seat near you when chairs rearrange. They wait for the side conversation rather than competing with the main one. The narrowing is rarely announced. It's enacted.
At a friend's birthday, sara was the social one — touching everyone's elbow, refilling drinks, introducing strangers. Halfway through the evening Mark realized she had ended up next to him for the third time. Each time it had looked accidental. Each time it had taken a quiet little maneuver — declining to follow someone to the kitchen, choosing the chair on his side of the couch. Friendly with everyone. Geographically loyal to him.
Mehrabian's immediacy concept (Mehrabian, 1971) captures this precisely: voluntary physical orientation toward a specific person is a high-signal cue, more reliable than verbal warmth. Reis and Shaver (1988) note that early intimacy is often built through small chosen proximities long before anyone names what's happening.
This week, in any group setting where you wonder, stop tracking the words and track the geometry. Who keeps ending up near you. Who turns their torso toward you when you speak. Who chooses you when there's a free choice — chairs, walking partners, conversation breakoffs. The room is talking even when nobody is, and the placements people quietly engineer are usually the most honest sentence of the evening.
Teasing is the great dividing line
Friendly people don't tease much. They're warm, they joke, but they avoid the slightly pointed, slightly personal needling that flirting uses as a calling card. Teasing requires a particular intimacy — enough attention to have noticed a specific quirk, enough trust to risk poking at it. When someone starts to tease you specifically, with affection rather than edge, they have already filed you in a different category than the rest of the room.
Mark mentioned offhand at lunch that he had recently bought running shoes. Two days later sara passed him in the hallway and said, oh look, it's the runner. A week after that, when he was carrying coffee instead of going for a lunchtime run, she said, what happened to the marathon dream. Nobody else in the office had remembered, let alone built a small private bit out of it. The teasing was a flag he was still learning to read.
Gottman's research on friendship and humor in couples (Gottman & Silver, 1999) treats gentle, specific teasing as a marker of attentive affection — distinct from sarcasm or contempt, which Gottman (1994) identifies as a top predictor of relational breakdown. The difference is whether the joke increases warmth and shared recognition or extracts it.
This week, notice who teases you, and notice the texture. Affectionate teasing leaves you slightly more visible and slightly more chosen. If it leaves you smaller, that's a different signal entirely. The first is often early flirting. The second is something to step away from.
Physical proximity tells the truth when everything else is ambiguous
When other cues are ambiguous — the eye contact is fine, the conversation is interesting, you genuinely can't tell — proximity tends to tell the truth. Bodies don't lie reliably about interest. A person who is merely friendly will keep a comfortable, polite distance even when the room allows them to close it. A person who is interested closes the distance gradually whenever the social cover allows: leaning in, shifting closer on a bench, ending up shoulder-to-shoulder somehow.
Sara and Mark had been talking on a slightly noisy patio for forty minutes. She had originally been across from him. Somewhere around the second drink she had pulled her chair around the corner of the table to hear better. Twenty minutes after that she was close enough that her knee was an inch from his. He had not asked her to move. Each adjustment had looked like it was about the noise. The trajectory was telling a different story.
Mehrabian (1971) treated proximity as one of the cleanest immediacy cues precisely because we tend to manage it semi-consciously. Eibl-Eibesfeldt (1972) noted that across cultures, courtship involves graduated closure of physical distance, usually with a plausible practical excuse for each step. The excuses vary. The pattern does not.
This week, when you can't read the situation, stop interpreting the words and watch the spatial drift over the next twenty minutes. Closing distance with plausible reasons is rarely accidental three times in a row. And if you notice yourself doing it, you've answered your own question about whether you're flirting back.
Nervous energy around you specifically is a tell
Some people are nervous in general. The cue is not nervousness itself; it's nervousness that appears specifically around you, in someone who is otherwise composed. Sudden self-consciousness, fidgeting with a glass, laughing a half-second too long at a small joke, a flash of hand-to-neck or hand-to-hair — these are not failures of confidence. They are signs that your specific presence has raised the stakes for them.
Mark watched a coworker run a tense team meeting calmly. Twenty minutes later in the kitchen she dropped a spoon, laughed at her own joke before she finished telling it, and tucked her hair behind her ear three times in two minutes. Nothing in the meeting had rattled her. Something about the four feet of empty kitchen between them was rattling her now. He didn't say anything. He just registered the difference and stopped wondering whether he'd been imagining the energy.
Mehrabian's work on nonverbal leakage (Mehrabian, 1971) is relevant: stress and arousal show up in the periphery of behavior even when the central performance is composed. Hess (1965) documented similar involuntary signals like pupil dilation in attraction contexts. Eibl-Eibesfeldt (1972) treated self-grooming gestures as widely-observed courtship tells across cultures.
This week, watch for the gap between someone's general composure and their composure around you specifically. If they are visibly steadier with the room than with you, that asymmetry is information. And if you notice yourself becoming clumsier around one specific person, ask yourself what that's telling you — about them, and about you.
The digital behavior extends the physical signals
Digital behavior tends to extend physical signals rather than contradict them. A friendly person likes your post when it appears in their feed. A flirting person watches every story within ten minutes of posting it, replies to specific details, references something from your feed in person the next day. The pattern isn't volume; it's specificity and timing. Generic engagement is friendly. Particular engagement, repeatedly, is something else.
Mark posted a story about a small concert he'd been to. A coworker replied that night: the second song is so underrated, did they play it. The next morning, in passing, she said, so was the second song actually live or did you make that up. He had told no one else about the concert. She had not just watched the story; she had remembered which song he'd called out, slept on it, and built a bit out of it the next day. The behavior was extending, not just acknowledging.
Reis and Shaver (1988) describe intimacy as the experience of specific responsiveness across contexts. Digital attention that crosses back into in-person reference does exactly this. Aron et al.'s closeness work (Aron et al., 1997) supports the broader principle: targeted attention that builds across moments is one of the most reliable accelerants of mutual interest.
This week, look at someone's digital engagement pattern with you over the last month. Is it broad and intermittent, or narrow and specific. Do they reference your posts in person. Do they reply to the details rather than just the headline. The pattern is usually clearer than any single message and harder to misinterpret than a single screenshot.
Pulling it together
The honest summary is that flirting is intentionally ambiguous, and trying to read it with complete certainty defeats the point of the behavior. What you can do is read the pattern more accurately than the cultural defaults allow, and treat the cluster of small selective behaviors as evidence rather than insisting on a smoking gun that flirting rarely produces by design. The ambiguity is part of the safety, and respecting that ambiguity is part of reading it well.
If there is one small change to try this week, it is to watch one familiar person across a social setting and notice how they behave with others compared to how they behave with you. The contrast, or the lack of it, is usually clarifying. People reveal who they are interested in mostly by what they do not do with everyone else.
When the cluster is selective and consistent, it is rarely just friendliness, and acting on that information honestly is almost always better than waiting indefinitely for a clearer sign that flirting was never going to provide.
Frequently asked questions
What if the person is just naturally flirty with everyone?
Then the absolute level of flirtation tells you very little, and the comparison becomes the relevant signal. Naturally flirty people still flirt selectively when they are actually interested; the markers shift from intensity to specificity. Watch for things that cannot easily be distributed across a room, such as remembered details, deeper questions, sustained attention, and follow-up that happens outside the social setting. The general warmth may be evenly spread, but the specific behaviors that require effort and memory are usually not, and that asymmetry is where the real signal lives.
Is teasing always a sign of flirting?
Usually a signal, but not always a romantic one. Teasing is essentially a low-stakes way of testing how someone handles slight social challenge, and that test is valuable in both friendship and flirtation. What distinguishes flirtatious teasing tends to be the topic, often appearance, the tone, often warmer than the words, and the selectivity, often disproportionate to one person in particular. Universal teasing, especially in established groups, is usually just the texture of the group. Targeted, slightly affectionate teasing directed mostly at you, especially when it is reciprocated, is more often a romantic register.
Can someone flirt without realizing they are doing it?
Yes, frequently, and this is one of the underrated complications. Many of the behaviors on this list are partly involuntary, particularly the proximity shifts, the eye contact patterns, and the small grooming gestures. People sometimes display the cluster without consciously deciding to, and may even deny the intent if asked directly. This does not necessarily mean the interest is not there; it sometimes means the conscious mind has not yet caught up to what the body has noticed. The pattern is still real even when it is not yet acknowledged.
What is the safest way to test whether it is flirting?
A small, low-cost escalation. Not a declaration, not a dramatic move, just a slight increase in attention or specificity that gives them room to either meet you there or gracefully redirect. The most useful tests are ones where a friendly response is still a comfortable place to land. Examples include suggesting a one-on-one activity that has a clear pretext, asking a slightly more personal question than the conversation otherwise required, or directly noticing something they have done that no one else seemed to. How they respond to the small move usually clarifies things faster than continued ambiguous reading.
Sources
What this article draws on. For the broader bibliography see our sources page.
- Mehrabian, A. (1971) — Silent Messages: Implicit Communication of Emotions and Attitudes
- Eibl-Eibesfeldt, I. (1972) — Similarities and differences between cultures in expressive movements (eyebrow flash research)
- Hess, E. H. (1965) — Attitude and pupil size, Scientific American
- Hazan, C. & Shaver, P. (1987) — Romantic love conceptualized as an attachment process, Journal of Personality and Social Psychology
- Fisher, H. (2004) — Why We Love: The Nature and Chemistry of Romantic Love