AI film · Guardian · 2026

The aisle ritual

A concept film for Guardian Express Click & Collect, made with generative AI. It names a small frustration that every beauty shopper knows, and then makes it disappear. This page covers the idea, the prompts, and the generation that failed and taught me the most.

Film · Click & Collect

Date night. One lipstick shade. Every shade but hers.

Client

Guardian Singapore

Product

Guardian Express Click & Collect

Status

Concept

My role

Concept, AI direction, edit and post

Tools

Seedance 2.0 on Higgsfield, Claude, After Effects

Year

2026

01

01

The idea

Insight

Health and beauty is exact-item shopping. You aren’t looking for a lipstick, you are looking for your shade, your SPF or your formula. So every store trip follows the same ritual. You look up, you look down, you crouch, and then you ask the staff, who offer to check the back. You already know how it ends. The fear isn’t that the store is out of stock. It’s that it’s out of yours.

Platform

Guardian, confirmed. It is the only way to shop that tells you the item is there before you go.

Proposition

Order on the app, and your item is confirmed, reserved and waiting at your chosen Guardian in as little as an hour. There is no ritual, no gamble and no wasted trip.

Big idea

Name the ritual, then end it. The film plays the aisle ritual out beat by beat, and every line is written in the second person, so each shopper sees themselves doing it. The answer isn’t a better search. It is never having to search at all: know it’s there before you are.

02

02

The script

There is no dialogue in the film. The story is carried by supers that speak directly to the viewer, one beat of the ritual at a time. That choice came out of a failed generation, which is covered further down the page.

Script
GUARDIAN EXPRESS CLICK & COLLECT: "THE AISLE RITUAL"

A woman in her early thirties searches a cosmetics shelf for one lipstick shade. We watch her from inside the shelf.

SUPER:  It's date night.
SUPER:  You need your shade.
SUPER:  You look up.
SUPER:  You look down.
SUPER:  You crouch.
SUPER:  Every shade but yours.
SUPER:  You'll ask the staff, but you'll already know how this goes.

CUT TO: the Guardian app, selecting a store.
SUPER:  Guardian Express Click & Collect. Collect in as fast as 1 hour.

END CARD: Know it's there before you are. Guardian.

Put the camera where no camera goes

The shot sits inside the shelf and looks out through the gap between two shelf levels. The shelf becomes a letterbox, and her search plays out inside it.

Lock the camera and move the person

The camera never moves. When she looks up, her face leaves the frame, and when she crouches, it drops back into the gap. The frame itself tells you what she is doing.

Keep the light ordinary

The prompt asks for flat, bright retail light and forbids a moody grade. Beautiful light would turn this into a lipstick ad instead of a familiar problem.

03

03

Character and location

The shopper is built the same way as the characters in my MSIG films: a locked reference sheet and a text description that goes into every prompt word for word. The store reference is used only for materials, fixtures and light, and the prompt tells the model not to copy its camera angle.

Character sheet of the shopper in front, three-quarter and profile views

The shopper. Front, three-quarter and profile views on one sheet.

Bright health and beauty store interior with pink display plinths

The store reference, used for materials, fixtures and light only.

Prompt: character sheet
Character reference sheet, three panels side by side on a seamless light grey backdrop, identical lighting and identical character across all panels.

CHARACTER: Singaporean Chinese woman, early 30s, working professional. Soft oval face, warm light-medium skin with natural texture, dark almond-shaped eyes, natural full brows, subtle everyday makeup with a muted berry lip. Long dark brown hair past the shoulders, loose soft waves, side part, tucked behind one ear. Slim, average height, relaxed confident posture.

WARDROBE: Deep burgundy sleeveless halter-neck knit top, high-waisted tailored black trousers, simple black block-heel mules. Thin gold hoop earrings, no other jewellery. Minimal, polished, office-to-dinner.

PANEL 1: Full body, front view, standing naturally, arms relaxed at sides, neutral expression.
PANEL 2: Waist-up, three-quarter view facing left, slight open-mouthed half-smile, mid-conversation, eyes engaged.
PANEL 3: Head-and-shoulders close-up, clean side profile facing right, calm neutral expression, earring and hair texture clearly visible.

LIGHTING & LOOK: Soft diffused key from front-left, gentle fill, no harsh shadows. Photorealistic, shot on cinema camera with 50mm lens, shallow depth of field, slightly desaturated filmic grade with soft greens and warm skin tones, fine grain. Consistent face, hair, wardrobe and proportions across all three panels.

NEGATIVE: no text, no labels, no logos, no props, no extra people, no background scenery, no heavy makeup, no glamour retouching, no plastic skin, no inconsistent outfits between panels.

04

04

Prompting

The video prompt describes a single twelve-second take with one continuous action. It spells out who is in the frame, where the camera is, what the lens does, what happens every two seconds, how the light should look and what should be heard. You can copy the full prompt below.

Prompt: the aisle ritual, 12 seconds
SCENE CONTEXT
Seen from inside a cosmetics shelf, looking out through the gap between two
shelf levels. A woman in her early thirties is already midway through
searching the shelf for one specific lipstick shade. She looks up, looks
down, crouches to eye level, takes one box, checks it, puts it back. It is
not hers.

ACTIVE REFERENCES
@shopper: early-30s Singaporean Chinese woman, long dark wavy hair worn
loose, gold hoop earrings, sleeveless dark burgundy knit top, black tailored
trousers, small handbag on her right shoulder. Dressed slightly smarter than
an ordinary work day. 100% matches the reference.

@store: the store interior. Use only for materials, fixtures, ceiling
structure, shelf construction and light quality. Do not copy its camera
angle or its wide composition.

@shopper is the only person in the foreground and midground, and the only
person whose face is legible at any point. Two or three shoppers may pass
far in the deep background, soft and out of focus. Every background person
is Singaporean — Chinese, Malay or Indian. No Western or Caucasian people
anywhere in frame. No staff member in frame. No duplicate of @shopper.

LOCATION MAP
The camera is physically inside a cosmetics shelf unit, resting on the shelf
surface, looking outward into the store aisle. The shelf above runs across
the top of frame as a dark horizontal bar. The shelf surface runs across the
bottom of frame. Between them is a wide letterbox gap through which the
aisle is visible.
Rows of small plain lipstick cartons stand in the immediate foreground on
both sides of the camera, close to the lens, out of focus, silhouetted
against the aisle light. All packaging is plain and unbranded. No readable
logos, no readable brand names, no recognisable retail chain identity
anywhere in frame.
Behind @shopper, across the aisle, the far side of the store is visible and
soft: pale display plinths, illuminated fixtures, the store's polished
floor.

FIRST FRAME AND SPATIAL BLOCKING
The first visible frame already contains @shopper, standing directly in
front of the shelf, 0.6 to 1 meter from the lens, centred in the letterbox
gap, already mid-search. She is settled in place from frame one — no
arrival, no walking in, no entering the frame. Her torso and the lower part
of her face fill the gap; her eyes sit just above the top shelf edge and are
partly cut off by it. No empty establishing frame. No delayed character
reveal.
Her body faces the shelf square-on, toward camera. Her gaze goes to the
shelves, not to the lens. She does not look at camera at any point.
Handbag stays on her right shoulder throughout.
Out-of-focus lipstick cartons occupy the left and right edges of frame for
the entire take, framing her through the shelf.

FORMAT MODE
Single continuous take. No cuts. No transitions. Real-time motion.
Total duration 12 seconds.

CAMERA
The camera is completely locked off. Tripod-rigid, fixed inside the shelf.
No pan, no tilt, no push, no handheld movement, no drift, no reframing of
any kind for the whole take. The frame never changes. @shopper moves within
a static frame.
Camera height is at the height of the shelf she is searching, roughly at
her chest when she stands and at her eye level when she crouches.

OPTICS
47° diagonal field of view, standard normal lens character, camera 0.6 to 1
meter from subject, natural human-eye perspective. Zero obvious distortion,
natural face and body proportions. The compressed letterbox look comes from
the shelf framing and the foreground occlusion, not from the lens.
Focus is held on @shopper. The lipstick cartons in the immediate foreground
fall soft. The far side of the store behind her falls soft.
No extreme wide distortion, no fisheye curve, no telephoto compression. The
lens character does not change at any point.

ACTION TIMING
0:00 to 0:02
@shopper is already standing at the shelf, mid-search. Her eyes move along
the row directly in front of her, small head movements, fingers of her left
hand resting lightly on the shelf edge. Nothing else in frame moves except,
far behind her, one soft out-of-focus shopper crossing the aisle.

0:02 to 0:04
She lifts her chin and looks up at the higher shelves. Her face rises out of
the letterbox gap; for two seconds we see only her neck, shoulders and chest
through the gap while she scans above.

0:04 to 0:06
Her chin drops. Her face returns into the gap and keeps travelling down,
eyes moving to the lower shelves.

0:06 to 0:08
She crouches in one controlled motion, knees bending, handbag sliding
forward on her shoulder. Her face comes fully into the gap and settles level
with the lens, close and clear, still looking at the shelves and not at
camera.

0:08 to 0:11
Her hand enters frame from the right, close to the lens. Her fingers take
one lipstick carton from the foreground row. She brings it toward her, turns
it to read the base, and holds it for a beat. Her face changes very slightly
— the corner of her mouth, a small drop in the eyes. She has her answer. She
reaches back in and sets the carton down in the row, not quite straight.

0:11 to 0:12
She does not move. She stays crouched, looking at the shelf. One breath.
Hold on her stillness until the shot ends.

PHYSICS
Real weight in every movement. The crouch has knee load, a small settle of
balance, and visible weight transfer. Hair shifts and settles a beat behind
her head movement. The knit top creases at the waist as she bends. The
handbag strap slides and pulls against her shoulder with real mass. The
carton has light cardboard weight — it tips slightly as she sets it down and
stays where she leaves it. The neighbouring cartons in the foreground are
not disturbed. No rubbery motion, no weightless props, no game-engine
physics.

LIGHTING
Two sources. Cool white LED strip light from inside the shelf unit, falling
on the foreground cartons from behind and above the lens, rimming their
edges. Bright overhead retail lighting in the aisle beyond, falling on her
from directly above, putting soft light on her forehead, cheekbones and
shoulders with faint shadow under her brow.
This flat commercial retail light is deliberate and correct. Do not
beautify it. No warm golden key, no cinematic moody grade, no rim light on
her face, no shaft of sunlight, no lens flare, no dramatic contrast. The
store stays clean, bright and ordinary.

AUDIO
Ambient store tone only: air conditioning hum, distant unintelligible
chatter, faint footsteps on polished floor, the light knock of the carton
being set back on the shelf.
No dialogue of any kind. @shopper does not speak. Her lips stay still for
the whole take. No off-screen voices. No narration, no voiceover, no music,
no subtitles, no captions, no on-screen text.

POSITIVE CONSTRAINTS
Kodak Vision3 250D, naturalistic documentary texture, real grain, grounded
physical cinema. Sharp clarity, natural colors, stable picture, no blur,
no ghosting, no flickering. All packaging unbranded and unreadable. No
brand logos anywhere in frame.

Decisions inside the prompt

One action, timed to the second

The search is broken into two-second beats: scanning, looking up, looking down, crouching, checking one carton and putting it back. Each beat has room to land.

Describe everyone in the frame

The prompt says she is the only person whose face is legible, that the few background shoppers are Singaporean, and that no staff member appears. Nothing is left for the model to guess.

Silence is written, not assumed

It states that she does not speak and her lips stay still. The story moved into supers, so no mouth has to match a line.

Ban the brand, not the look

All packaging is plain and unreadable, so no real product or competitor shows up on the shelf.

05

05

What didn’t work

My first attempt tried to fit the whole story into ten seconds. The camera followed her down the aisle, she crouched and went through the ritual, and then she turned to a staff member who said, “Let me check the back.” It didn’t work, and the way it failed shaped the final film.

The failed generation. Watch the shopper’s mouth when the staff member speaks.

Too much action for the time

A follow shot, a crouch, the search and a conversation were all packed into one ten-second clip. Every beat felt rushed and none of them landed.

Both characters said the line

The prompt didn’t make clear enough who speaks, so the shopper and the staff member both mouthed “Let me check the back”.

The shopper drifted

She came out as a Western woman. I hadn’t pinned down her ethnicity firmly enough, so the model filled the gap with its default.

The fix was to do less. The final film uses one locked shot, one action and twelve seconds. The staff member and the dialogue were cut, and the words moved into supers. Her ethnicity, and the ethnicity of everyone in the background, is now stated in the character sheet and again in the video prompt.

06

06

What I took away

01

Give each shot one action. When a model has too much to do in ten seconds, it does all of it badly.

02

If a line of dialogue could belong to two people, the model will often give it to both.

03

Anything you don’t specify gets filled with the model’s default, so describe every person in the frame, including the ones in the background.

04

A failure can improve the idea. Moving the words into supers turned an ordinary scene into a direct address to the viewer.

Concept, AI direction, edit and post by Farhan Haniff. A concept for Guardian Singapore.