Mini VSL

Mini VSL je skraćena verzija video prodajnog pisma: od trideset do devedeset sekundi, sa punom prodajnom strukturom. Ne prodaje benefitom nego mehanizmom, odnosno novim razumevanjem problema koje gledalac nije imao. Zato radi za skuplje proizvode i usluge gde jedna rečenica nije dovoljna da se opravda cena.

Write concept, script and explanations in the user's language. **Write all generation prompts in English**, because the models perform better and the user pastes them unchanged.

What it earns, honestly

Hook rate će biti niži i to je u redu. Duži format gubi više ljudi na početku, i to je namerno. Ne meri VSL istim aršinom kao yapper. Ono što se meri je koliko ljudi stigne do poslednje trećine i šta oni rade posle. Ako je hook rate nizak a konverzija onih koji ostanu visoka, format radi kako treba.

Why it persuades

• Duža forma filtrira. Ko odgleda devedeset sekundi je ozbiljno zainteresovan.
• Mehanizam menja poređenje. Kad kupac razume zašto tvoje rešenje radi drugačije, prestaje da poredi po ceni.
• Struktura je dokazana decenijama u direktnom marketingu i ne zavisi od trenda.
• Jedan dobar VSL nosi ceo nalog mesecima.

Stage 1 — Interrogate the product

Ask what the user is selling, at what price, to whom, and what the single claim is. Do not proceed on assumptions. If they already gave enough context, answer for them and show it for confirmation rather than asking again.

The governing rule: the concept is derived FROM the product, never applied TO it. Quality test before delivering: would this same concept work with a competitor's product? If yes, rewrite.

Stage 2 — Pick the style variant

VSL prodaje ozbiljnošću i mehanizmom. Glineni ili crtani stil ukida ozbiljnost baš u trenutku kad tražiš da neko izdvoji veći novac. Izuzetak je varijanta B, gde animirani prikaz mehanizma pomaže razumevanju, ali narator mora da ostane ljudski.

If the user does not choose, recommend one and say why in one line.

Stage 3 — Script the beats

Mehanizam je srce formata. Bez njega je ovo dugačka reklama, a ne VSL.

Write the actual lines verbatim in the user's language. Do not describe what is said.

Stage 4 — Five hooks against one body

Always deliver five opening variants. When hold rate is fine and hook rate is low, only the first three seconds get regenerated, which is five ads for the price of one production.

Structures that hold for this format:

• „Ako imaš [problem] i probao si sve, evo zašto ništa nije radilo.“
• „Za devedeset sekundi ću ti pokazati zašto...“
• Brojka koja odmah kvalifikuje ili diskvalifikuje
• „Ovo nije za tebe ako...“
• Priznanje greške pre nego što padne ijedna tvrdnja

The hook must be a single visible thing. "Curiosity" is not a hook.

Stage 5 — Production

Order of work, do not deviate

1. Napiši ceo skript pre ijedne slike VSL pada ili prolazi na tekstu. Vizuel ga samo prati.
2. Podeli skript na blokove od deset sekundi Svaki blok je jedan kadar ili jedna grafika.
3. Generiši govornika i zaključaj ga Kod dužeg formata drift je zagarantovan bez reference.
4. Generiši prikaze mehanizma Isti stil kroz sve, kao u objašnjivaču.
5. Sklopi i titluj ceo tekst Reč po reč. Duži format se najviše gleda bez zvuka.

Model routing

Shot	Image	Video
Govornik, srpski	Nano Banana Pro	Google Omni Flash
Govornik, engleski	Nano Banana Pro	Veo 3.1
Prikaz mehanizma	Nano Banana Pro	Kling 2.5 ili Seedance 2.0
Naracija bez lica	nema slike	Omni Flash za glas, montira se preko

On-camera Serbian goes through Google Omni Flash. Other models carry a foreign accent local audiences detect within half a second. If pronunciation still fails, generate the voice separately and apply lip sync with an audio-driven tool over the finished image rather than accepting a foreign-sounding take.

Locked parameters

Every prompt carries these, explicitly or by clear implication. Changing one means generating a different format.

Prompt skeletons

KADAR, govornik (Nano Banana Pro)

[PERSON DESCRIPTION] framed from the chest up, seated or standing in [ORDINARY PROFESSIONAL SPACE], looking directly into the lens with eyes locked to the lens, already mid-gesture. The space behind is real and lived-in with [TWO SPECIFIC OBJECTS] visible but thrown out of focus. One soft key light from camera left plus gentle fill. Static camera at eye level, 50mm. Natural skin texture with visible pores, neutral grade, no cinematic look. No logos, no readable text.

**KADAR, mehanizam** (Nano Banana Pro)

A clean stylised diagram of [MECHANISM] on a plain [COLOUR] background, showing [THE THREE PARTS], with [THE KEY PART] highlighted in [ACCENT COLOUR] while everything else stays desaturated. Flat even lighting, no shadows, clean neutral grade. Educational illustration quality, not photoreal, not cartoon. At most five elements in frame, no text.

**VIDEO, govor** (Omni Flash za srpski)

[FIRST FRAME] speaks. In [LANGUAGE], measured and deliberate, as if explaining something important to one person, with one small natural stumble: "[LINE]". The head moves naturally with the speech, hands entering the bottom of frame occasionally. Eyes stay locked to the lens throughout. The camera does not move. Audio: close voice, ordinary room acoustics, no music. A single uninterrupted take filmed in one breath.

### Two phrases that are not optional

Every video prompt closes with `A single uninterrupted take filmed in one breath`, or the models
insert hidden cuts. Never write `zoom`; write `the camera travels toward`. Never name a phone or
camera, or the model draws one into the shot.

---

## Worked example — Proizvod: konsalting za automatizaciju, 1.200 evra. Mehanizam: gubitak vremena nije u zadacima nego u prelazima između njih.

**PRIMER · KADAR** (Nano Banana Pro)

A woman in her forties in a plain dark shirt seated at a desk in an ordinary office, framed from the chest up, looking directly into the lens with eyes locked to the lens, already mid-gesture with one hand raised. Behind her a shelf with folders and a closed laptop visible but thrown out of focus. One soft key light from camera left plus gentle fill. Static camera at eye level, 50mm. Natural skin texture with visible pores, neutral grade, no cinematic look. No logos, no readable text.

**PRIMER · VIDEO** (Omni Flash)

The frame speaks. In Serbian, measured and deliberate, as if explaining something important to one person, with one small natural stumble: "Vaš tim ne gubi vreme na zadatke. Gubi ga na... na prelaze između njih, i to se nigde ne meri." The head moves naturally with the speech, one hand entering the bottom of frame. Eyes stay locked to the lens throughout. The camera does not move. Audio: close voice, ordinary room acoustics, no music. A single uninterrupted take filmed in one breath.

Use this to calibrate how much detail belongs in each bracketed slot.

---

## Failures and repairs

**Nema mehanizma**
→ Bez novog razumevanja problema ovo je duga reklama, ne VSL. Vrati se na skript.

**Govornik se menja kroz kadrove**
→ Kod dužeg formata drift je zagarantovan. Reference Element je obavezan.

**Tempo previše spor**
→ Sporiji od kratkih formata ne znači sa pauzama. Iseci svaku pauzu u montaži.

**Ponuda je nejasna**
→ Cena i sledeći korak moraju da budu izgovoreni, ne samo napisani.

**Lažna hitnost**
→ Rok koji ne postoji ruši poverenje i krši pravila. Ako roka nema, ne izmišljaj ga.

**Hook rate je nizak**
→ To je normalno za format. Meri koliko ljudi stigne do poslednje trećine.


**Three-attempt rule.** If the same problem survives two generations, change the approach rather
than the prompt: add a reference image, halve the shot, or split the action into two generations
joined in the edit.

## Assembly

- **Redosled** Hook, problem, mehanizam, dokaz, ponuda. Od trideset do devedeset sekundi.
- **Rezovi** Između blokova, na kraju misli. Unutar bloka jump rezovi na pauzama.
- **Titlovi** Ceo tekst, reč po reč. Bez izuzetka.
- **Zvuk** Muzika tiho ispod, staje na ponudi.
- **Verzije** Napravi verziju od trideset i verziju od devedeset sekundi. Testiraj obe.

---

## Deliver

KONCEPT:      [one paragraph, what happens and why it comes from this product]
VARIJANTA:    [which style, and why]
SCENARIO:     [beat by beat, with verbatim lines]
PET HOOKOVA:  [five openings against the same body]

Then every prompt in English, complete and pasteable, labelled by shot. Then one line on what to
check first in the result.

## Never

- Deliver prompts without a concept behind them.
- Deliver a concept that would survive a product swap.
- Return prompt fragments or advice about what the user should edit. Full prompts only.
- Recommend a style variant this format does not support.
- Let the word "zoom" or a named recording device into a prompt.
- Substitute a model when Serbian speech is on camera.