What a Long-Form YouTube Video Really Costs to Make

"How much does a video cost" is usually answered with a number that hides the real answer. The real answer is a list of tasks, each with an hour count, and the number falls out of it.

Here is that list for a 15-minute long-form explainer — the format most channels and most product teams actually publish.

The task list

Every long-form video, however it is made, involves the same nine jobs:

1. Idea and angle — deciding what the video argues, not just what it is about
2. Research — sources, facts, numbers, the specifics that make it credible
3. Scripting — 2,000 to 2,500 words for 15 minutes, structured for retention
4. Script review — the pass that fixes pacing and cuts what does not earn its place
5. Visual planning — what appears on screen, scene by scene
6. Visual production — footage, illustration, motion, or generation
7. Narration — recording or generating, plus cleanup
8. Assembly — cutting picture to audio, pacing, transitions
9. Packaging — title, thumbnail, description, chapters

Nothing on that list is optional. The only question is who does each one and how long it takes.

The traditional numbers

Doing it all yourself. Realistically 20 to 35 hours for a well-made 15-minute video, and the variance is mostly in research and assembly. At any honest valuation of your own time this is the most expensive option on the list — it just does not send an invoice.

Hiring it out. In most markets, per video: scripting 150–500, voiceover 100–300, editing 300–1,200, thumbnail 30–150. So roughly 600 to 2,000 per video, plus your own time briefing, reviewing, and re-briefing — which is never zero and is usually 3 to 6 hours.

In-house. A part-time editor plus tools lands somewhere in the same range once you amortise salary across output, with better consistency and worse flexibility.

The number that matters is not the total. It is the cost per attempt — because the hard part of YouTube is not making one video, it is finding out which of your ideas works.

Why cost per attempt is the real metric

Channels are found by iteration. You do not know which format, angle, or topic will land until you have published enough to see a pattern, and that usually takes twenty-plus videos.

At 1,000 per video, twenty attempts is 20,000 and most people never run the experiment. At 30 hours per video, twenty attempts is fifteen weeks of full-time work. Both prices are high enough that creators stop early — and stopping early is the actual failure mode, not any individual video being bad.

Anything that lowers cost per attempt lowers the price of finding out you were right.

Where an AI pipeline changes the numbers

It does not change the task list. It changes who does which job.

Automatable: script drafting, visual production, narration, assembly, rendering. These are the hours, and they are the invoices.

Not automatable: idea and angle, fact checking, script review, packaging. These are the judgement, and they are what makes the video yours.

For a 15-minute video, that shakes out to roughly:

• Brief and direction — 10 minutes
• Script review, scene by scene — 25 minutes
• Theme and voice — 5 minutes
• Final edit pass and watch-through — 20 minutes
• Packaging — 20 minutes

About 80 minutes of attention, plus generation cost. In credits at Prompt2YT's flat rate, a 15-minute video is 75 credits for the video itself plus visuals — single-digit dollars, not four figures.

What you should still spend money on

Cutting production cost is not the same as cutting quality, and there are two places where spending still pays:

Research. If your video makes claims, buying the book, paying for the archive, or interviewing someone who was there is what makes it worth watching rather than a summary of the top three search results.

Packaging. A great thumbnail designer earns their fee back on impressions alone. This is the one task worth outsourcing even when everything else is automated, because it is high-leverage and genuinely hard.

The honest summary

• A traditionally-produced 15-minute video costs 600–2,000, or 20–35 hours of your own time
• The same video through an AI pipeline costs roughly 80 minutes of attention plus single-digit generation cost
• The saved money is best redirected into research and packaging, not pocketed
• The saved time is what lets you run twenty attempts instead of three, which is the only reliable way anyone finds a format that works