How to choose, and then stop.

Why your own face is the hardest subject.

There is a mechanical reason this feels harder than it should. You know your face as a reflection, reversed, and a photograph does not reverse it. Every small asymmetry you long ago filed under normal turns up on the wrong side, and your eye reads it as wrong rather than as unfamiliar. Add the ordinary strangeness of seeing yourself still, at a size and for a length of time you never otherwise do, and a first look through a gallery is rarely a fair trial.

The way out is not to become more objective about your own face, which nobody manages. It is to stop judging the face and start judging the photograph against the job it has to do. Nobody opening your profile is auditing your jawline. They are working out, quickly, whether you look like a person who does the thing your headline says you do.

So the question that unlocks a gallery is never which of these do I like. It is which of these does the work. What separates the two is mostly craft rather than luck, and what makes a photo look expensive takes that apart element by element.

The first pass: instinct, at speed.

Open the gallery and go through it once at a walking pace. Mark anything that makes you feel like yourself. No enlarging, no comparing, no notes, no counting. If an image makes you pause, mark it. If you have to think about it, do not.

• Go through the entire set once before you go back to anything.
• Do not enlarge on this pass. Enlarging finds pores, not photographs.
• Mark generously. A wide shortlist costs nothing and narrows fast.
• Do it alone, and not at the end of a day when you are already worn down.

Speed is the whole point of this pass. Instinct reads an expression well, and deliberation is where self-criticism gets in and starts renegotiating your face. You are only building a shortlist here. Nothing you mark is a commitment, and nothing you skip is gone.

The second pass: against the list.

The thinking pass starts on paper rather than on screen. Write down the places these photographs genuinely have to work, in the order of how much each one matters to you.

• The profile somebody opens before a meeting with you.
• Your own site, where the picture sits beside your words.
• The photograph and paragraph you send when an organiser asks for both.
• Anything printed: a program, a page, a wall.

Now put each marked image against one of those lines. Some fit two lines. Some fit none, and those come off the shortlist however fond of them you are. The crop each destination wants is not the same, which is why a photograph can be lovely and still wrong for the slot it is going into. The full inventory of destinations is in where to use your new headshot everywhere.

If any line on your list involves ink, flag it now rather than later. An image chosen on a bright laptop can sit heavier on paper, because paper reflects light instead of emitting it, and screen and print color never agree perfectly.

One image, one job.

One image, one job, is the discipline that ends the paralysis. A gallery does not need a single winner. It needs an answer for each line on your list, and those answers are allowed to be different photographs.

Naming the job also settles the practical questions attached to it. A profile picture will be small and often circular, so it lives or dies on the eyes and the expression, and the background barely registers. A picture for a site header will be wide and large, so it needs room beside you that a headshot crop does not contain. A portrait for print has to survive being enlarged, which is a different question again, and how much resolution you need settles that one.

Write the job beside each choice in a single word. Profile. Site. Bio. Print. The decision is then defensible for a reason that has nothing to do with your chin, which is exactly what you want at the moment somebody questions it.

The one you love that does not fit.

Every gallery contains this trap. It is the image you adore that cannot do the job. Usually you look softest in it, or youngest, or most relaxed, and usually you are looking away, or laughing with your eyes half closed, or lit so gently that at small size your features dissolve into a pleasant blur.

Test it instead of arguing with yourself. Shrink it to about the size of a fingernail and look at it from arm's length. If you can still read the expression, it survives and belongs on the list. If it turns into an agreeable smudge of hair and jacket, it is a photograph for your family rather than for your profile, which is a fine thing for it to be.

The opposite trap is just as real. The strongest working picture is often the one you liked least on the first pass, because in it you are looking straight down the lens and being seen, which is uncomfortable to look at and effective everywhere else. Direct is not the same as severe. If the expression is honest, keep it in contention even while your instinct objects.

The second opinion, and who to ask for it.

Ask exactly one person, and pick them deliberately. Somebody who works with you will judge the photograph against the job. Somebody who loves you will judge it against the person they see walking through the door, and their answer, however kind, is about their memory of you rather than about the picture's function.

Ask a specific question, too. Which one is best produces a shrug and a compliment. Try which of these three looks most like me in a meeting, or if you had never met me, which of these would you rather call. Specific questions get specific answers, and specific answers move you along.

• One person, not a group message. Committees converge on the blandest option.
• Show two or three images, never the whole gallery.
• Show them at the size the picture will really be used, not full screen.
• Ask about the expression, not about the shirt.

How to stop, and what comes next.

Stopping is a decision rather than a feeling that arrives. Set the limit before you begin: a shortlist, one second opinion, a night to sit with it, then done. Longer is not more accurate. Past a point everything starts to look strange, including your own name written down.

Once you have your answers, do the practical part in one sitting: download the whole set at full size, keep the untouched originals somewhere safe, and put the working copies where you will find them again. The routine for that is in what you get with your digital files, and it takes less time than the choosing did.

And if you are stuck between two, call and ask me. I have looked at every frame, and which of two images carries better at small size is a craft question rather than a matter of taste. That is the same conversation that decides what a sitting should produce in the first place: headshots covers the work itself, and the investment page explains why the numbers happen on the phone rather than on a page.

For the whole subject from the beginning, including how the frames you are choosing between came to exist, the complete guide to professional headshots runs start to finish.