How to Learn Guitar from Scratch
The same short-session approach applied to another skill that rewards daily practice far more than it rewards equipment.
Read the guide →
The gap between a snapshot and a photograph almost never comes down to equipment. It comes down to four decisions made before the shutter: where the light is coming from, what is in the frame and what is left out, what is sharp, and how bright the result should be. Modern phones handle the technical work well enough that those four decisions are now the whole game. Over six short sessions this guide works through them one at a time, with exercises you do on whatever camera you already own — phone, compact or system camera. You will end with a small set of habits you apply before pressing the shutter rather than a list of settings to memorise.
What you'll be able to do: In six short sessions, learn to see light, frame deliberately and control focus — so the camera you already own starts producing photographs instead of snapshots.
Outcome: "Before I shoot, I notice where the light is coming from and how hard it is."
Look at any person or object near you and work out, in five seconds, which direction the light is hitting it from. Do that ten times today.
Photography is recording light, so the first question is always where it comes from and how hard it is. Direction decides shape: light from the side creates the shadows that make a face or a landscape look three-dimensional, light from directly in front flattens everything, and light from behind puts your subject in silhouette unless you compensate. Quality decides mood: a small, distant source — midday sun, a bare bulb — is hard light with sharp shadow edges and high contrast, while a large, close source — an overcast sky, a big window, sun through a curtain — is soft light that wraps around and flatters. The single most useful rule follows from this: an overcast day is not bad light, it is a softbox the size of the sky, and it is easier to shoot in than bright noon sun. If the light is bad, the cheapest fix is almost always to move — around the subject, or to a different time of day. Changing camera changes nothing about this.
A coffee cup on a table, photographed four times in twenty minutes. Backlit against a window: the cup is a dark shape against a blown-out white rectangle, and the steam glows — dramatic, but the cup itself is lost. Turned so the window is to the left: the curve of the cup gets a bright edge and a soft shadow, and it suddenly reads as a solid object. Moved to the middle of the room under the ceiling light: flat, dull, slightly orange, no shape. Back beside the window but with a sheet of white paper propped on the right: the paper bounces light back into the shadow, the contrast drops, and the cup looks like a product shot. Four photographs, one subject, one camera, no settings changed — only the light.
Outcome: "Every photo I take has one clear subject and nothing competing with it."
Take a photo you already have and crop it so it contains only what you actually wanted. Notice how much was in there that you never meant to include.
Composition is mostly subtraction. Beginners' photos are usually not badly arranged; they simply contain too much, so the eye has nothing to settle on. Decide the one subject, then remove or reposition everything competing with it. Three tools do most of the work. Get closer — the commonest fix in all of photography, and physically moving beats zooming, especially on a phone. Check the edges and the background before shooting, not after: the lamppost growing out of a head is always visible in the viewfinder and never noticed there. Use lines — paths, fences, shadows, railings — that lead the eye toward the subject. The rule of thirds, which puts the subject a third of the way in rather than dead centre, is a useful default rather than a law; turn on your camera's grid and use it until off-centre framing becomes instinct, then break it deliberately. One more habit worth building: keep horizons level. A tilted horizon reads as a mistake in a way that almost nothing else does.
A market stall on a busy street. First frame, taken from where the photographer happened to be standing: the stall occupies a third of the picture, a parked van fills the left edge, a road sign cuts the top corner, and a passer-by's back blocks the produce. No subject. Second frame, three metres closer and crouched slightly: the van is out, the sign is out, the stall fills the frame and the vendor's hands are visible — but the stall sits dead centre and the background is a cluttered shopfront. Third frame, two steps to the left so the shopfront is replaced by a plain shuttered wall, vendor placed a third of the way in, the line of the awning running down toward him. Same stall, same camera, ninety seconds apart. The difference is entirely in what was excluded.
Outcome: "I choose what the camera focuses on, and I can make a background blur or stay sharp."
Open your camera, tap somewhere far away, then tap something close. Watch what happens to the rest of the frame. That is depth of field, demonstrated in ten seconds.
Two separate ideas often get confused. Focus is where the sharp plane sits, and on any camera you should be choosing it rather than letting the camera guess — tap the subject on a phone, or use a single focus point on a system camera. For portraits, the focus goes on the near eye, always; an in-focus nose with soft eyes reads as a mistake to everyone, even people who cannot say why. Depth of field is how much in front of and behind that plane is acceptably sharp, and three things control it: a wider aperture (a smaller f-number) blurs more, getting closer to the subject blurs more, and pushing the background further away blurs more. On a phone, the first is largely fixed, which is why the other two matter so much — the most effective way to get a blurred background on a phone is to move the subject away from it and step in closer, not to switch on portrait mode. Portrait mode simulates the effect and gives itself away around hair and glasses, so use it, but get the distances right first.
Subject standing with her back against a hedge, phone at arm's length, portrait mode on. The result: background blurred, but the fake blur cuts into her hair on the left and leaves a hard halo around one shoulder. Second attempt, portrait mode off: she moves four metres forward, away from the hedge, and the photographer steps in until she fills the frame from the shoulders up, tapping her near eye to focus. The hedge, now four metres behind her and out of the focal plane, goes genuinely soft — no halo, because nothing was simulated. Third attempt, portrait mode back on with the new distances: slightly more blur than the second, and this time the effect has less work to do, so the edges hold up. The order matters — distance first, software second.
Outcome: "I adjust exposure deliberately instead of accepting whatever the camera decides."
Open your camera, tap a subject, then slide the little sun icon up and down. Most phone users have never touched it.
Your camera's meter aims for an average mid-tone, which means it is systematically wrong in two common situations: it underexposes a bright scene like snow or a white wall, making it grey, and it overexposes a dark one, making it muddy. Every camera offers exposure compensation — the slider that appears next to the focus square on a phone, or the +/− dial on a system camera — and using it is the single biggest improvement available in a menu. The asymmetry worth knowing: a shadow that is too dark can usually be lifted later, but a highlight that has gone pure white contains no information at all and cannot be recovered. So when in doubt with a bright sky or a white shirt, expose a little darker and lift afterwards. If your camera can show a histogram, the useful reading is simple: data piled hard against the right edge means blown highlights. Finally, shoot RAW if your camera offers it and you intend to edit — it holds far more recoverable detail than JPEG, at the cost of larger files.
A person standing on a ridge with a bright overcast sky behind. Tapping the face to focus makes the camera expose for the face, which is much darker than the sky — the face looks correct and the entire sky turns to featureless white. Sliding exposure down by about one stop brings the sky back as actual cloud with texture, and the face goes too dark. That second version is the one to keep, because the face can be lifted in editing while the white sky in the first version is empty data. Third attempt: the photographer moves so the person is against the hillside instead of the sky, which removes the problem rather than managing it. That is usually the better answer — exposure compensation handles the situations you cannot move out of.
Outcome: "I can direct someone simply and catch them between poses instead of in one."
Ask someone to stand still and look at the camera. Then ask them to walk toward you and stop. Compare the two faces.
The hard part of photographing people is not technical. A camera pointed at someone makes them self-conscious, and self-consciousness is visible in every frame — the fixed smile, the locked shoulders. Three things dissolve it. Give a task, not a pose: walking, pouring, looking at something, doing the thing they are good at. Bodies doing something look natural in a way that bodies being photographed do not. Keep shooting past the moment — the frame you want is usually the one after the posed one, when they relax because they think you have finished. Talk, because silence behind a camera is what makes people brace. Technically, keep it simple: put them in soft light, ideally beside a window or in open shade rather than direct sun, which makes people squint and casts hard shadows across the eyes. Shoot from around their eye level or slightly above — from below is rarely flattering — and give a specific small direction rather than "act natural", which is the least actionable instruction in existence. And ask before photographing strangers; in much of Europe, publishing a recognisable photo of someone taken in public can need their consent.
Someone who says they hate being photographed, in a kitchen. First frames, posed against the wall: shoulders up, a smile that arrives and leaves on command, eyes slightly wide. The photographer moves her to the window, gives her a mug to hold, and asks her to look out at the garden and describe what the neighbours have done to the fence. She starts talking. Her hands do something, her shoulders drop, and her face does five different things in thirty seconds. The keeper is a half-smile mid-sentence with soft window light across one side of her face — a frame she was not aware of. The technical settings never changed between the bad frames and the good one.
Outcome: "I can improve a photo in five adjustments and pick the few worth keeping."
Open your best photo from this week and move only the exposure slider, nothing else. Stop when it looks right, and notice how small the move was.
Editing has two parts, and the second matters more. The first is adjustment, and five controls handle almost everything: exposure for overall brightness, contrast for punch, highlights and shadows to recover the bright and dark ends separately, white balance to fix a colour cast, and a crop to finish the composition and straighten the horizon. That is the whole toolkit for most photographs. The failure mode is obvious in other people's work and invisible in your own: saturation pushed until skin goes orange, shadows lifted until the picture turns grey, clarity raised until everything has a halo. A useful discipline is to edit, walk away, and look again the next day; the overdone version always announces itself overnight. The second part is selection, and it is what separates a portfolio from a camera roll. From a hundred frames, keep five. Showing someone your three best photographs makes you a better photographer in their eyes than showing them forty, of which three are good.
A window-light portrait, shot deliberately a little dark to protect the bright window behind. In the editor: exposure up a third of a stop, so the face reads correctly. Shadows lifted slightly to open the dark side of the face — slightly, because pushing this one is what makes photos look washed out. Highlights pulled down to bring back texture in the window frame. White balance nudged warmer, because the overcast light outside had left a faint blue cast on the skin. Finally a crop: a little off the left edge where a radiator intrudes, and a two-degree rotation to level the windowsill. Five moves, about ninety seconds, no filter, no saturation. Then the harder step — of the thirty-one frames from that session, three are kept and the rest deleted, including four that are perfectly fine and simply not as good.
The complete guide — all 6 chapters, every checklist and exercise — as a clean, printable PDF. No email required.
A one-page cheat sheet of the six chapters, a 75–90 minute photo-walk script to run this live with someone else, and 12 affirmations and reflections to keep nearby.
Every guide on this site is produced with the Self-Directed Learning Framework: instead of an article that explains a topic at you, the framework turns it into a structure you work through — each chapter with one outcome, a starting action under twelve minutes, the minimum you actually need to know, a worked example, a checklist, and exercises. It starts from three questions:
Topic: Taking better photographs with an ordinary camera or phone.
Outcome wanted: Make deliberate choices about light, framing, focus and exposure before shooting, and finish with a small edited set — within six days.
Starting point: Takes photos regularly on a phone, likes some of them, and cannot say why those ones worked.
The prompt framework behind it is public: see the C.R.A.F.T. prompt framework, including the exact prompt that generated this library.
The same short-session approach applied to another skill that rewards daily practice far more than it rewards equipment.
Read the guide →Free, self-paced guides across everyday, professional and academic topics.
See all guides