How to cut a cooking tutorial that covers every step and lands on the plate

Last updated 2026-09-10 · 8 min read

What is the short version?

Kitchen footage is recorded in the order the cooking happened, which is not the order a tutorial needs, and it always has a gap — one step where the hands were in the way or the camera had been nudged.

1. Go to Vizard Agent with the footage and the steps.
2. Ask which steps have no usable shot.
3. Ask for the cut to end on the serving moment.

What do you need before you start?

The footage and the recipe written as stages. Vizard Agent maps one against the other, and without the list it can only cut what looks good — which is how tutorials end up missing the step everybody actually needed to see.

• All the footage. Including the unpromising clips.
• The recipe, as steps. Prep, cook, fold, plate, taste.
• Which steps matter most. Some are the whole video.
• The target length. It decides how many stages fit.
• Whether you can reshoot. A gap may be fillable.

What do you type into Vizard Agent?

Ask about coverage before you ask for a cut. The useful first answer is not an edit — it is a list of which steps you have and which you do not, because that is the only thing you can still do something about.

Prompt

Variants worth knowing:

• "Which steps have no clear shot." The question that matters.
• "The strongest moment for each stage." One per step.
• "Finish on the first taste." Names the ending.

What does Vizard Agent actually do?

Here is the sequence on a real cooking tutorial cut from a single session's footage. The coverage map comes first, which is what turns "here is some footage" into a decision about what can be made from it.

1. Builds a contact sheet of everything that was filmed.
2. Reviews it to understand the shot coverage.
3. Maps the ingredients, the steps and the payoff against it.
4. Reports which steps have no usable shot.
5. Reviews each stage's clips for the clearest beat.
6. Picks the strongest prep shot, and one for each stage after it.
7. Chooses a strong shot for the middle, where tutorials sag.
8. Checks the inserts for burnt-in bars or framing problems.
9. Builds the cut in recipe order, not filming order.
10. Ends on the plating and the taste.

Step four is the step that earns its place. Finding out during the edit that the folding step was never filmed cleanly is bad; finding out before anything is cut means you can reshoot one thirty-second clip while the pan is still out.

Step seven is about attention rather than instruction. The middle of a tutorial is where people leave, and it is usually where the least interesting stage sits — so it needs the best available shot rather than the one that happens to be next.

Step ten is the difference between a tutorial and a recording. Footage stops when somebody stopped filming; a tutorial ends on the result, and the taste is the shot that makes a viewer want to cook the thing.

What does the result look like?

A cut that follows the recipe, shows every stage you actually have, and finishes on the dish. Vizard Agent also tells you which stages it could not cover, which is what you need for the next shoot as much as for this one.

When does this not work well?

An edit cannot film what was not filmed. If a step is genuinely missing, the options are a reshoot, a still, or a caption that carries it — and no amount of clever cutting turns three unusable clips into a step.

• A step that was never filmed. Reshoot, or caption it.
• Hands blocking the action. Common and hard to fix.
• Very long recipes. They do not fit a short.
• One continuous take. There is nothing to select between.
• No payoff shot. The ending has to come from somewhere.

How do you fix a result that came back wrong?

Refer to the recipe steps rather than to clips or timecodes. Because Vizard Agent built the cut against the recipe in the first place, a note like "the folding step is unclear" points straight at a stage and the shots available for it rather than at a position on the timeline.

• "The folding step is unclear." A different shot for that stage.
• "It drags in the middle." A stronger shot, or fewer stages.
• "You missed the seasoning." It may have no usable footage.
• "End on the plate, not the pan." The payoff shot changes.

How does Vizard Agent compare to doing it yourself?

By hand the footage gets watched and the nice-looking shots get used, which produces a good-looking video that quietly skips a step. The gap is found by a viewer trying to follow along, which is the worst possible time.

Common questions

Do I need to write the steps out? It helps enormously. Vizard Agent maps coverage against them.

What if a step was not filmed? Vizard Agent tells you before cutting, while a reshoot is still cheap.

Can it use a still for a missing step? Yes, or a caption. Vizard Agent treats both as better than skipping it silently.

Does it cut in cooking order? Vizard Agent cuts in recipe order, which is not always the same thing.

How long should each step be? Long enough to follow. Vizard Agent sizes them to the target length.

What about the narration? Vizard Agent can write it to what the hands are doing.

Will it keep the kitchen sound? Yes if you want it. Vizard Agent will point out the sizzle is worth keeping.

What if hands block the action? Vizard Agent will look for a clearer moment in the same clip.

Can it do a longer version too? Yes. Vizard Agent builds it from the same coverage map.

Does the middle really matter? It is where people leave, so Vizard Agent puts a strong shot there.

What if there is no taste shot? Vizard Agent will say so; the ending needs one from somewhere.

Can it add step numbers on screen? Yes. Vizard Agent ties them to the stages it mapped.

Will it use every clip? No, and Vizard Agent will tell you what it left out and why.

Why do tutorials skip steps? Because they are cut from the best-looking footage rather than against the recipe.