About the service
I cut vertical video for social feeds: you send the footage, I send back clips that are ready to publish. The first pass is always selection. I watch the whole source, mark the moments where the delivery is clean and the thought reaches its end, and set the rest aside before a timeline even exists. A short vertical clip forgives nothing, so the choice of fragments decides far more than any effect added later.
Then comes rhythm. The opening seconds carry the entire clip, so it starts on a strong line rather than on a greeting, and every pause that holds nothing is removed. Subtitles are generated automatically and then read through by hand: names, terms and numbers are exactly the places where recognition slips. Captions and accents go on top, and deliberately few of them, because text covering half the screen is read by nobody. Colour is matched between shots, loudness is brought to one level so the voice does not jump from clip to clip, and the cover is built around a headline that stays legible at thumbnail size.
Three things I keep to without exception. Music comes only from libraries that grant commercial use, and someone else's track is never dropped under your video. The meaning of what you said is not reversed in the edit: a phrase torn out of context does not go into a clip. People in frame who did not agree to publication are blurred or cut out.
— going through the whole source and picking the fragments that work
— edit rhythm built to hold attention in the first seconds
— automatic subtitles with manual correction of the errors
— captions and accents placed sparingly, without clutter
— colour correction between shots and levelled audio
— a cover with a headline that reads at small size
— separate versions for different platforms
— a package of several clips made from one shoot
— the editing project handed over to you after delivery
Then comes rhythm. The opening seconds carry the entire clip, so it starts on a strong line rather than on a greeting, and every pause that holds nothing is removed. Subtitles are generated automatically and then read through by hand: names, terms and numbers are exactly the places where recognition slips. Captions and accents go on top, and deliberately few of them, because text covering half the screen is read by nobody. Colour is matched between shots, loudness is brought to one level so the voice does not jump from clip to clip, and the cover is built around a headline that stays legible at thumbnail size.
Three things I keep to without exception. Music comes only from libraries that grant commercial use, and someone else's track is never dropped under your video. The meaning of what you said is not reversed in the edit: a phrase torn out of context does not go into a clip. People in frame who did not agree to publication are blurred or cut out.
— going through the whole source and picking the fragments that work
— edit rhythm built to hold attention in the first seconds
— automatic subtitles with manual correction of the errors
— captions and accents placed sparingly, without clutter
— colour correction between shots and levelled audio
— a cover with a headline that reads at small size
— separate versions for different platforms
— a package of several clips made from one shoot
— the editing project handed over to you after delivery
