Most privacy advice arrives as a list of things to switch on. Use a password manager. Enable two-factor authentication. Cover your webcam. All of it is reasonable, and all of it answers a question you never asked. A threat model flips the order: first decide what you are protecting and from whom, then pick the handful of measures that match. The whole exercise takes about half an hour with a pen and a sheet of paper, and it will save you from years of security theater aimed at the wrong problem.
A threat model is a short written answer to three questions: what do I have that is worth protecting, who might want it, and what would it realistically take for them to get it? It is not an audit and it does not need to be complete. It is a filter.
Once you have one, every recommendation you meet gets the same test: does this help against a risk on my list? If yes, do it. If no, skip it without guilt. Declining a tool that solves someone else's problem is not laziness, it is judgment.
Two myths are worth setting aside. You do not need to be interesting to be targeted, because automated attacks do not choose victims, they scan for them. And "I have nothing to hide" misses the point, since the assets at stake are usually access, money, and attention rather than secrets.
Start concrete. "My privacy" is too vague to defend. Write down the things that would cause real damage if they were taken, exposed, or locked away from you.
Then rank them with one blunt test: if I lost this tomorrow and never got it back, what breaks first? The top three are your priorities. Ignore the rest for now.
"Hackers" is not an adversary, it is a costume. Useful threat modeling describes capability and motive. Most people face a small set of realistic actors:
Targeted state-level surveillance is real, but for most readers it sits far down the list. If your work or activism puts you in that category, get advice specific to your situation rather than stretching a generic checklist to cover it.
Now combine the two lists into sentences. Each one should name a specific asset and a plausible route an adversary could take to reach it.
Vague: "Someone might hack my accounts." Useful: "A criminal buys a password dump, finds my old reused password, logs into my email, and from there resets my bank login."
Write six or eight of these, then sort them into three buckets: likely and damaging, possible but survivable, and not worth the effort. The first bucket gets your attention this week. The second gets cheap fixes when you have an hour spare. The third gets accepted, consciously, which is different from being ignored.
Common entries on real threat models include a phone with no screen lock and an unlocked mail app, a router still using its default admin password, photo backups that quietly hold scans of ID documents and screenshots of one-time codes, and one shared password reused across streaming, shopping, and email.
If you cannot describe how the attack would actually work, you are not threat modeling yet. You are worrying.
Good defenses are cheap, permanent, and cover several scenarios at once. In rough order of return:
Write your chosen actions down with today's date. Keep the list short enough that you would genuinely do it.
A threat model ages. A new job, a new phone, a move, a breakup, a parent who needs help with their accounts: each one changes what you have and who might want it. Put a reminder in your calendar twice a year and spend fifteen minutes rereading your notes. Cross off what you fixed and add whatever changed.
If you are dealing with a stalker, an abusive ex, or any situation where someone with technical skill and personal knowledge is targeting you, skip the generic version and contact a specialist support organization in your country. Threat modeling helps, but it is not a substitute for expert help when your safety is on the line.
Photo: StartupStockPhotos / Pixabay
