Midjourney

Verified against Midjourney · 2026-07-21

Get a genuinely clutter-free product shot using real negative prompting

An e-commerce product-photography brief that leans on --no, Midjourney's real negative-prompting mechanism, to reliably strip out props, extra units, and background clutter that a positive-only description often fails to keep out.

Midjourney v78 fillable variables

The prompt

Ready to copy — highlighted parts are example details you can swap.

PRODUCT
a matte-black ceramic pour-over coffee dripper with a walnut wood collar

PLACEMENT AND SURFACE
centered on a honed light-grey concrete surface

BACKGROUND
a seamless soft off-white gradient background fading to pale grey at the edges

LIGHTING SETUP
a large soft overhead softbox key light with gentle fill from the left, subtle rim light separating the product edge from the background

CAMERA ANGLE
three-quarter angle, slightly above eye level, sharp focus on the ceramic glaze texture and wood grain.

RAW MODE FOR LITERAL COLOR
--style raw — a listing photo needs to represent the product's actual color and finish, not Midjourney's default tendency to push a "pleasing photo" toward warmer, more saturated tones than the real item.

NEGATIVE PROMPTING
--no text, watermark, second unit of the product, hands, price tag, cardboard box — this is Midjourney's genuine negative-conditioning mechanism, distinct from describing what should be in frame; use it specifically for the recurring failure modes of product shots — a second unit of the product appearing in the background, a hand or prop nobody asked for, text or a watermark-like mark — rather than trying to prevent all of them purely through positive description alone.

PARAMETERS
--style raw --no text, watermark, second unit of the product, hands, price tag, cardboard box --ar 1:1 --v 7

OUTPUT
a matte-black ceramic pour-over coffee dripper with a walnut wood collar, centered on a honed light-grey concrete surface, a seamless soft off-white gradient background fading to pale grey at the edges, a large soft overhead softbox key light with gentle fill from the left, subtle rim light separating the product edge from the background, three-quarter angle, slightly above eye level, sharp focus on the ceramic glaze texture and wood grain --style raw --no text, watermark, second unit of the product, hands, price tag, cardboard box --ar 1:1 --v 7

WHAT --NO ACTUALLY REMOVES VERSUS WHAT IT DOESN'T
--no reliably suppresses concrete, nameable things — an object category, text, a color. It does not reliably suppress a vague quality like "clutter" stated as a single word, because the model has nothing concrete to subtract for an abstract term; if the background still reads as busy after a first attempt, replace a vague --no clutter with the specific objects that actually showed up unwanted (--no cardboard box, price tag, second product) rather than assuming a broader or repeated vague term will do more.

IF A SECOND UNIT OF THE PRODUCT KEEPS APPEARING
This is common enough in product photography prompts to name explicitly: add the literal phrase describing a duplicate — "second unit," "duplicate item," "multiple products" — to the --no list, since Midjourney's default composition instinct for product-category scenes sometimes includes more than one unit unless told not to.

Customize

Optional — swap in your own details for the highlighted parts above.

Why this works

--no is Midjourney's real negative-conditioning mechanism, not a phrase the model might choose to honor loosely — unlike Flux or Nano Banana, which have no negative-prompt field at all and must steer unwanted content out purely through positive description, Midjourney's underlying process supports true negative conditioning, which is why it is the more reliable tool here for the specific, recurring failure modes of product photography: a second unit appearing in the background, an unrequested prop, watermark-like text. The distinction drawn between concrete, nameable exclusions and vague quality words is the practical difference between --no actually working and --no doing nothing measurable: a term like "clutter" has no specific referent for the model to subtract, since it names a subjective quality rather than an object category, while "cardboard box" or "price tag" names something concrete the model can specifically suppress. A user who tries --no clutter, doesn't get a cleaner result, and concludes negative prompting doesn't work for product shots has usually just used it on the wrong kind of term — the fix is naming the actual unwanted object once it's visible in a first attempt, not abandoning --no for a purely positive-description approach. Pairing --no with --style raw addresses a separate, unrelated default bias that --no cannot touch: --no controls what's absent from the frame, while raw mode controls how faithfully the present content's color and finish are rendered. Without raw mode, Midjourney's aesthetic layer tends to push product colors toward a warmer, more saturated "pleasing photo" grade that can misrepresent the product's actual color to a buyer — a defect --no has no ability to fix since it's not about removing an unwanted object, it's about the fidelity of a wanted one. The explicit callout for duplicate-product suppression reflects a specific, well-observed pattern in this exact use case: Midjourney's learned association for a product-category scene sometimes includes rendering more than one unit of the product, likely because many training images of a product category legitimately show multiples (a shelf of items, a set), and a single-product listing shot needs to explicitly override that association rather than assume "one product" was implied by only describing one in the positive prompt text.

What you get back

A single, cleanly-lit product photo on a seamless background with true-to-life color, no second unit, no stray prop, and no watermark-like text — ready to use as a listing image with minimal or no further retouching.

Verified against

Midjourney v7 · 2026-07-21

Changelog

  • 2026-07-21 Initial publish, verified against Midjourney v7 --no negative prompting for clean product photography.

Need this built into your business?

If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.

EXPLORE WHAT SCULT BUILDS
All Midjourney prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY