Tutorials Video

Video Production Fundamentals: The Complete Guide

Beginner · ~35 min

Overview

Almost every video problem people ask about, muddy audio, footage that will not cut together, an export that looks worse than the timeline, was created earlier than the stage where it was noticed. Video production is a chain of constraints: what you decide in planning limits what the camera can do, what the camera records limits what the edit can fix, and what the edit produces limits what delivery can preserve. This guide walks the whole chain in order and explains what each stage locks in, so you can recognise which decision actually caused the problem you are looking at. It is the map. The linked guides are the territory.

What You Need

A camera that lets you control shutter, aperture, and ISO manually. A phone with a pro mode counts
Any microphone that is not the one built into the camera
A way to keep the camera still: tripod, gimbal, or a solid surface
One controllable light source, even if it is a window with a curtain
Editing software. DaVinci Resolve and CapCut both have capable free tiers
Enough storage for roughly three times your finished runtime in raw footage

Steps

Decide what the video is for before you decide how it looks

Every later choice resolves faster once the purpose is explicit. A tutorial that must be followed on a phone in a workshop wants close framing, high contrast, and loud clear speech. A brand film wants mood and can afford ambiguity. Write one sentence naming the audience, the platform, and the single thing the viewer should do or understand at the end. Almost every argument about a shot is really an unresolved argument about this sentence.

Plan in shots, not in scenes

A scene is a unit of story. A shot is a unit of work. Break what you need into a list of individual shots with a rough duration for each, and you get a schedule, a shot list, and a gear list almost for free. The list also exposes coverage gaps while they are still cheap to fix. You will notice you have no cutaway for the middle section now, on paper, rather than in the edit three days later.

Set exposure deliberately, then stop moving it

Shutter speed near double your frame rate gives natural motion blur. Aperture sets how much is in focus. ISO is the last resort because it costs noise. Lock white balance to a number rather than auto, or your footage will drift in colour between takes and refuse to match in the grade. Auto modes are not worse because they are inaccurate. They are worse because they change their mind mid-take, and inconsistency is far harder to fix than a mediocre constant setting.

Light for shape, not for brightness

A bright, evenly lit subject reads as flat and cheap. Depth comes from a dominant key light placed off to one side, enough fill to keep the shadow readable, and separation between the subject and whatever is behind them. Cameras handle low light better every year, so the reason to add light is almost never exposure. It is control over where the eye goes.

Record sound as a separate discipline

Audiences forgive a soft image and abandon bad audio within seconds. Get the microphone close, proximity beats microphone price by a wide margin, and listen on headphones while recording, because you cannot fix what you did not hear. Aim for peaks around -12 dBFS to leave headroom, and record thirty seconds of the room with nobody talking, which gives your editor a noise profile and a bed to patch cuts with.

Shoot coverage, not just the shot you pictured

The edit needs somewhere to cut. For any sequence longer than a few seconds, get a wide to establish, a medium for the substance, a close-up for emphasis, and at least one cutaway of something else entirely. Coverage is what lets you shorten, reorder, or rescue a take later. It is the cheapest insurance in production and the most common thing beginners skip.

Edit for structure first, polish second

Assemble the whole thing badly before you make any of it good. A rough cut that runs end to end tells you what is missing, what drags, and what can be cut entirely, and it prevents the classic failure of spending an afternoon perfecting an opening that gets deleted. Colour, sound sweetening, and graphics all come after the structure stops moving.

Export to the platform, not to a habit

Match your export frame rate to your shooting frame rate, use the codec and container the destination actually wants, and give the encoder enough bitrate that its own compression is not fighting yours. Most "my export looks worse" complaints are a bitrate set too low, a colour space mismatch, or a platform re-encode that a higher-quality upload would have survived.

Pro Tips

Fix problems as far upstream as you can. Ten seconds moving a microphone beats an hour of noise reduction, and moving a lamp beats a mask in the grade.
Shoot a few seconds of handles before and after every take. Editors need overlap to make transitions work, and you cannot add it later.
Slate every take, even alone, clap once on camera. It gives you a sync point and an audible take marker in one gesture.
Back up before you sleep, to two places, one of them off-site. Card corruption is rare until the one time it is not.
Watch your rough cut with the sound off, then with the picture off. Each pass exposes weaknesses the other hides.
If a shot needs explaining before people understand it, the problem is the shot, not the audience.

Knowledge Base

What You'll Learn

This guide is the connective tissue for the video tutorials on this site. Below is how the stages relate to each other, which mistakes propagate furthest downstream, and where to go next for depth on any single stage.

Why the order of operations matters

Video production is often taught as a set of independent skills (lighting, audio, editing) and that framing hides the thing that actually makes productions fail. The stages are not independent. They form a pipeline in which each stage can only ever degrade or preserve what the previous one handed it.

An editor can trim a long take but cannot invent a reverse angle that was never shot. A colourist can adjust contrast but cannot recover a blown-out window that clipped to pure white in camera. A mixer can reduce room reverb slightly but cannot un-reverberate a voice recorded from across a hall. This is why experienced crews appear to obsess over things that seem minor at the time: they are not being fussy, they are protecting options that vanish permanently at the end of each stage.

The practical consequence is that when something goes wrong, you should look upstream from where you noticed it. Footage that will not cut together is usually a coverage problem from the shoot, not an editing problem. A grade that will not match across a scene is usually an auto-white-balance problem from the camera. Diagnosing at the wrong stage is how people spend hours fixing symptoms.

The three things worth spending money on first

Gear advice tends to be either aspirational or defensive, and neither is useful when you are deciding what to actually buy. Ranked by how much visible difference they make per unit of cost, the order is consistently: audio, support, light, and the camera comes fourth.

Audio first because it has the steepest quality cliff and the least tolerance from viewers. A modest microphone placed close to a subject will beat an expensive one placed far away, so spend on getting the microphone into position (a boom arm, a stand, a lavalier) as much as on the microphone itself.

Support second because a stable image reads as competent almost independently of anything else, and a tripod is cheap relative to how much it changes perceived production value. Light third because control over contrast and direction is what separates footage that looks shot from footage that looks captured.

The camera is fourth because modern cameras, including phones, are past the point where the sensor is the limiting factor for most work. The limiting factor is nearly always the person operating it and the conditions they put it in.

Frame rate, shutter, and the one rule worth memorising

Frame rate is how many images per second you record. Shutter speed is how long each of those images is exposed for. They are independent settings that people conflate constantly.

The rule worth memorising is that shutter speed should sit at roughly double the frame rate, 1/50th of a second at 24 or 25 fps, 1/60th at 30 fps. This produces the amount of motion blur audiences read as normal, because it approximates the 180-degree shutter that film cameras used mechanically for a century. Shoot at a much faster shutter and motion becomes stuttery and hyper-crisp. Much slower and it smears.

Higher frame rates exist for slow motion. If you want footage to play back at half speed, shoot it at double the project frame rate and keep the shutter rule relative to the capture rate. The common mistake is shooting everything at 60 fps "for flexibility" and then delivering it at 60 fps, which gives everything the fluid, immediate look associated with live broadcast, fine for sport, wrong for almost anything intended to feel considered.

Where beginners actually lose quality

Across the tutorials on this site, the same handful of upstream mistakes account for most of the downstream complaints:

Automatic settings that drift. Auto exposure and auto white balance re-evaluate continuously, so a cloud passing the window changes your footage mid-sentence. The resulting inconsistency cannot be graded out.

Microphone distance. Sound intensity falls off sharply with distance while room reflections do not, so doubling the distance does not just make speech quieter. It makes the room proportionally louder. This is why on-camera microphones sound hollow.

No coverage. A single unbroken take gives the edit nowhere to go. Any change of timing then requires a visible jump cut.

Mismatched project settings. Editing 25 fps footage in a 30 fps timeline introduces judder that no amount of rendering will remove. Set the timeline to match the footage before you start cutting.

How to use the rest of this site

Treat this page as an index with reasoning attached. For the shoot itself, start with the guides on shooting B-roll and lighting a talking head on a budget, then move to depth of field, focus pulling, and exposure monitoring once the basics are automatic. For sound, the audio recording fundamentals guide is the parallel pillar to this one.

For post, work through the DaVinci Resolve beginner material before colour grading, since grading decisions are much easier once the interface is not also new. Export settings and the platform delivery guides close the loop.

The interactive tools are worth using while you plan rather than after: the exposure triangle simulator, the depth-of-field calculator, and the lens field-of-view tool all answer questions that are cheaper to resolve before you are standing on location.

FAQ

Q: Do I need an expensive camera to start?
A: No. Camera bodies passed the point of being the main quality limiter some years ago, and a current phone in a manual mode outperforms professional cameras from a decade ago. Spend on audio, stable support, and controllable light before upgrading a body. Those three change results far more per unit of cost.

Q: What frame rate should I shoot at?
A: Match the convention of your delivery region and intent: 24 or 25 fps reads as cinematic and is standard for narrative and brand work, 30 fps is common for online and corporate video, and 50 or 60 fps suits sport, gaming, and anything you intend to slow down. Whatever you pick, keep shutter speed near double the frame rate and set your timeline to match your footage.

Q: Why does my footage look worse after uploading?
A: Platforms re-encode everything you upload, and a file that was already heavily compressed degrades much further through a second pass. Export at a generous bitrate in the codec the platform recommends, and confirm your colour space is correct. An unflagged log or wide-gamut file will look washed out after processing.

Q: How much footage should I plan to shoot?
A: For scripted or structured work, roughly three times your finished runtime is a reasonable starting ratio. For documentary or interview work it is far higher, sometimes twenty to one. The number matters less than the coverage: it is better to have three angles of the important moment than twenty minutes of a single unusable wide.

Q: Should I fix problems in camera or in post?
A: In camera, wherever there is a choice. Post can only work with information that was recorded, so anything clipped, out of focus, or drowned in room reverb is permanently degraded. The exceptions are deliberate: shooting flat or log profiles intentionally defers colour decisions to the grade, which is a choice to keep options open rather than a repair.

Q: What is the single most common beginner mistake?
A: Leaving the microphone on the camera. It is the one decision that guarantees distant, room-dominated sound regardless of every other thing you do well, and audiences tolerate poor audio far less than they tolerate a poor image.