Skip to content

Building confidence one conversation at a time

Confidence in conversation is built with reps, not mindset. A five-chat plan with one small goal per chat, and the research on why it works.

LustAI Editorial5 min read

A row of paper-craft dumbbells shaped like speech bubbles, each one slightly larger than the last, on a plain surface

This article is for general information and is not medical or psychological advice. If something here touches a hard place for you, talking to a professional is a good next step.

Confidence isn't a mood you wait for. It's a record of things you've done, and the quickest way to add to it is a handful of short conversations with a goal each. Below is a five-chat plan, why reps beat pep talks, and a way to score yourself that doesn't depend on whether the other person liked you.

Why doesn't "just be confident" work?

Because it has the order backwards. In 1977, psychologist Albert Bandura published a theory of self-efficacy: the belief that you can handle a specific situation. His central finding was that the strongest source of that belief is what he called mastery experience, meaning you did the thing, it went okay, and you noticed. Being told you can do it helps a little. Watching others do it helps a little more. Doing it yourself, in small doses that succeed, is what changes the number.

That's why reading ten articles about confidence leaves you exactly where you started, and why one slightly clumsy chat that ends fine does more than all ten. Your brain doesn't update on intentions. It updates on evidence.

What does the research say about practising with strangers?

Two findings make strangers the perfect training partners. First, the fear is mostly a forecast error. When Sandstrom, Boothby and Cooney (2022) had people talk to at least one stranger a day for a week, using a scavenger-hunt app to nudge them, the participants became less afraid of being rejected, more confident in their conversational ability, and more likely to keep starting conversations after the study ended. One week. No coaching, just reps.

Second, you're being graded more kindly than you think. In Boothby and colleagues' 2018 "liking gap" studies, pairs of strangers chatted and then privately rated each other. Across every version, people believed their partner liked them less than the partner actually reported. The gap was widest for the people who felt most awkward. The awkwardness you can feel from the inside barely registers from the outside.

A sphere and its reflection. The reflection is calmer and softer than the original, which is roughly how the other person sees you compared with how you see yourself mid-conversation.

The five-chat plan

Five short conversations, over a week or a weekend, each with one goal you fully control. You "pass" a chat the moment you do the goal, regardless of how the other person responds. That last part is the whole trick: you're training the behaviour, not chasing approval.

  1. Chat 1: open first. Your only job is to send the first message. You could say: "Quick question, are you a morning person or is this the wrong time to ask?" Whatever comes back, you've passed. The conversation starters that work with strangers piece has more if you want options.
  2. Chat 2: ask one follow-up. When they say anything about themselves, ask one more question about it. Huang and colleagues (2017) found that people who ask more follow-up questions are liked more, in both lab conversations and speed dates. "You said you moved last year, where from?" counts. Follow-up questions are a skill of their own and worth ten minutes.
  3. Chat 3: say a real opinion. Not a controversial one. A true one. "I genuinely think pineapple on pizza is fine and I'll defend it." The goal is to exist on the page as a person with edges, not a mirror.
  4. Chat 4: let a silence sit. When the reply is slow, don't fill it with a second message for at least two minutes. Confidence in text is partly the ability to leave space. If two minutes feels like agony, that's the rep.
  5. Chat 5: end it well. Close the conversation yourself, kindly, before it fizzles. You could say: "I should go, but this was a nice ten minutes. Hope your week's decent." Ending on purpose is what turns a chat from something that happened to you into something you did.

How do I keep score?

Keep a plain rep log. It's the record Bandura was talking about, made visible. Rate two things after each chat on a 0 to 10 scale: how nervous you were before, and how awkward it felt during. Then note whether you hit the goal. Here's what a typical first week looks like.

ChatGoalNerves beforeAwkward duringGoal hit
1Open first87Yes
2One follow-up76Yes
3A real opinion66Yes
4Let a silence sit57Yes (barely)
5End it well44Yes

Notice the "awkward during" column doesn't fall in a straight line; chat 4 bumped back up because the goal was harder. That's normal. The "nerves before" column is the one that matters, and it almost always falls steadily, because each completed chat is evidence for the next one.

What about being funny, or charming, or interesting?

Later, and less than you think. Epley and Schroeder's commuter studies found that people who struck up conversations with strangers enjoyed themselves more than expected, and that the strangers were happy to talk. Nobody in those studies was performing. They were asking ordinary questions and listening to ordinary answers.

Humour helps, but it's mostly a by-product of being relaxed, and relaxed comes from reps. If you're curious about the mechanics, why funny works explains it without turning you into someone who rehearses jokes.

When it isn't working

If the nerves-before number hasn't moved after ten or more chats, check three things. Are the goals actually small? "Be interesting" isn't a goal; "ask one follow-up" is. Are you doing them, or reading about them? And are you quietly using a crutch, like drafting every message three times, that stops the rep from counting?

If you also notice that the fear of being judged is shaping choices well beyond chats, such as turning down invitations or avoiding calls at work, that may be more than practice can fix on its own. Shyness and social anxiety look similar from the outside and need different responses, and social anxiety and online chat covers how to use text as a first rung rather than a hiding place. There's more on this in the wellbeing section.

When to get more help

If conversation feels impossible rather than uncomfortable, or the dread has lasted months and is costing you work, study or friendships, a doctor or counsellor can help, and asking is itself the first rep.

Questions people ask

How do I become more confident in conversation?
Do more short conversations with a single small goal each, and count the goal as the win. Psychologists call the result self-efficacy: the belief you can handle something, built from having handled it. Reading about confidence doesn't build it. Five small chats do.
How do I stop being awkward in conversation?
Mostly by finding out that you're less awkward than you feel. Research on the liking gap shows people rate their conversation partners more warmly than those partners expect. Ask one more question than feels natural, and let silences be a second longer than you'd like.
Can I practise conversation skills online?
Yes, and it's a good gym for it: low stakes, endless partners, and a clear end. Use text to practise questions and follow-ups, voice to practise tone, and video to practise being seen. Keep sessions short and stop after you hit the goal.
How do I be more confident texting?
Send messages without re-reading them more than once, ask direct questions, and stop apologising for taking up space. A confident text is short and specific. The re-reading habit is what makes you feel unsure, not the message.
How long does it take to feel confident talking to people?
The fear of starting drops noticeably within a week of daily small chats in the research, and it keeps dropping with reps. Feeling fully at ease takes longer and never quite reaches zero for anyone. The goal is workable, not fearless.

Sources

  1. Bandura (1977). Self-efficacy: toward a unifying theory of behavioral change · Psychological Review (PubMed)
  2. Sandstrom, Boothby & Cooney (2022). Talking to strangers: a week-long intervention reduces psychological barriers to social connection · Journal of Experimental Social Psychology
  3. Boothby et al. (2018). The liking gap in conversations: do people like us more than we think? · Psychological Science (PubMed)
  4. Huang et al. (2017). It doesn't hurt to ask: question-asking increases liking · Journal of Personality and Social Psychology (PubMed)
  5. Epley & Schroeder (2014). Mistakenly seeking solitude · Journal of Experimental Psychology: General

Written by LustAI Editorial. We write plain-language guides about talking to people online, staying safe and feeling good about it. Every piece is researched against primary sources, edited by a human, and reviewed again whenever we update it.