One of the best things I've learned in my clinical training is that everything is information. Everything a client does provides data, and I can use it to generate hypotheses, even before the first session. For example, it's telling when a client asks the receptionist four times how much longer he'll have to wait, even though I'm only ten minutes behind schedule. It's interesting and useful to notice that a middle-aged client is dressed like a teenager, or that a client keeps his coat and hat on when he sits down across from me.
What can I hypothesize about these clients? The client who repeatedly checks with the receptionist might be anxious, entitled, or both. The middle-aged client dressed like an adolescent might be emotionally immature or fear aging. The client who keeps his coat and hat on might be emotionally guarded and unsure about seeing a psychologist. These are only possibilities, but they allow me to start generating a conceptualization of the client--one that will be reinforced or corrected during the interview.
Naturally, the habit of observation extends outside the therapy office. I notice if an acquaintance is usually on her phone when she enters a party, and I wonder if it's important to her to look busy or if she's socially anxious about greeting people. I notice if a colleague starts every sentence with an apology and try to guess if it's just a verbal tic or if he believes that his contributions to conversation have no value or that others aren't interested in his opinion.
You don't have to be a psychologist to notice behaviour and make hypotheses, and there are benefits to this kind of observation. Say a friend tells you that she's thrilled about her new job, but you notice that she's biting her nails and doesn't smile once during the conversation. You can gently point out that she seem more stressed out than thrilled, and give her a chance to reflect and to discuss her true feelings. If you notice that a friend often busies himself with drinks and hors d'oeuvres and spends most of his time in the kitchen during weekly games night, you can hypothesize that he's unsure of his place in the social group, and--without necessarily saying anything--make a particular effort to include him.
Next time a friend or stranger does something interesting or unusual, or you notice a pattern in a colleague's behaviour, make a hypothesis. You might learn something interesting!