Vizard Agent

How to cut property clips into a listing video

Last updated 2026-08-16 · 5 min read

Upload your property clips to Vizard Agent and say what the video is for. Vizard Agent inspects every clip, works out what each one shows and where its best few seconds are, then orders them into a tour that reads like a walkthrough rather than a folder of files. No sorting or trimming is needed first.

What is the short version?

A property shoot produces a dozen clips of rooms, and the editing job is almost entirely selection: which four seconds of the kitchen clip, and does the balcony come before or after the bedroom. That is judgement applied repeatedly, which is exactly what makes it slow by hand.

  1. Go to Vizard Agent and upload all the clips at once.
  2. Say what the video is for and roughly how long it should be.
  3. Say whether you want captions, music or neither.

What do you need before you start?

The clips as they came off the camera, and a sentence about the audience. Vizard Agent looks at every clip before deciding anything, so pre-sorting is wasted effort — but it cannot know whether this is a luxury listing or a rental ad, and that changes the pacing more than any other instruction.

What do you type into Vizard Agent?

Say what it is for rather than how to cut it. Vizard Agent handles the selection and ordering; what it needs from you is the purpose, because a video meant to sell a family home is paced differently from one meant to fill a rental quickly.

Prompt

Edit these clips together into a converting video to sell a house. It is for a real estate marketing agency. [No subtitles.] Keep it under [30] seconds.

Variants worth knowing:

What does Vizard Agent actually do?

Here is the order Vizard Agent worked in on a real nine-clip property shoot. Almost all of it is looking: first at the files, then at a montage of the frames, then at what happens inside each individual clip as it plays — which is the only way to know that the good part of the kitchen shot is the second half.

  1. Probes all nine clips for resolution, duration and format.
  2. Pulls a thumbnail from each and stitches them into one montage, then looks at it to see what rooms it has been given.
  3. Checks rotation metadata on the clips, because phone footage stores its orientation separately from its picture.
  4. Checks whether the clips carry audio, then runs speech recognition on the first one to see whether anyone is narrating.
  5. Builds frame-by-frame grids for individual clips to trace camera motion, door openings and reveals.
  6. Reads each grid in turn — the door in clip 1, the reveal in clip 4, the pacing of the kitchen in clip 9, the balcony slider in clip 3 — to pick the best segment of each.
  7. Assembles the tour from those chosen segments.

Step 6 is the whole job. Vizard Agent is not sampling a frame per clip; it is watching what happens inside each one and choosing where it starts and stops.

What does the result look like?

From the run this page is written from, probed on the delivered file: 1080x1920, H.264, 30fps, 27.3 seconds, AAC audio. Nine clips came back as one vertical tour under half a minute, which is the shape most property posts want.

Ask for a second version and Vizard Agent re-cuts from the same analysis, so a wide version for the listing page costs a fraction of the first pass.

How long it takes was not measured separately here. Comparable work runs a median of 28 to 38 minutes end to end, and the upper quartile is several times the median. Across all projects the median cost by tier is Flash 47, Pro 55, Max 242, Ultra 263 credits.

When does this not work well?

The limits here are the limits of the footage. Vizard Agent chooses well among what you shot and cannot choose what you did not shoot, and property walkthroughs have a characteristic set of problems that no amount of editing repairs — most of them are notes for the next viewing rather than for this video.

How do you fix a result that came back wrong?

Say which room and what is wrong with it. Vizard Agent keeps the clip analysis, the chosen segments and the assembled cut, so swapping a section or reordering the tour does not mean re-inspecting the footage — the expensive part has already been done.

How does Vizard Agent compare to doing it yourself?

By hand this is import nine clips, scrub each one to find its good four seconds, trim, order, and export. None of it needs skill. All of it needs attention, and the attention is per clip, which is why a twelve-clip shoot takes longer than a nine-clip one in a way that feels unfair.

By hand Vizard Agent
Reviewing footage Scrub every clip Probes and inspects all of them
Choosing segments Watch, mark in and out Reads frame grids per clip
Ordering the tour Arrange by instinct Builds a walkthrough order
Second version Re-edit for the other shape Re-cut from the same analysis

Common questions

Do I need to sort the clips first? No. Vizard Agent looks at all of them and works out what each one shows.

Can it add music? Yes. Say the mood, or say no music if the listing is going somewhere that plays muted.

Can I get a vertical and a wide version? Yes. Ask for both and the second is a re-frame rather than a fresh edit.

Can it handle clips shot on different phones? Yes. Vizard Agent probes each clip separately, so mixed resolutions and frame rates are normal input rather than a problem to solve before uploading.

Will it add captions with the property details? It will if you give it the details. Vizard Agent does not know the asking price or the floor area unless you say.