How to create product and marketing images with image2img
A practical guide to creating product images, ad creatives, social posts, and website visuals with prompts, reference images, and saved image workflows in image2img.
How to create product and marketing images with image2img
image2img is built for people who need practical visual assets, not just one-off AI art experiments. You can use it to create product images, ad creatives, social posts, portraits, landing page graphics, and visual variations from a prompt or a reference image.
This guide walks through a simple workflow that matches how image2img is used today: describe the image, choose the right format, add a reference when it helps, generate inside your workspace, then save or continue from the result.
Start with the final use case
A good image prompt starts with where the result will be used. A product listing image needs clarity and accurate composition. An ad creative needs stronger contrast and a clear subject. A website graphic may need negative space for text. A portrait may need a specific mood, lens, and lighting direction.
Before writing the prompt, decide:
- What is the image for?
- Who or what is the main subject?
- What format do you need: square, portrait, landscape, or wide?
- Should the result look realistic, editorial, clean, cinematic, illustrated, or stylized?
- Do you need the output to follow an existing product, face, layout, or brand direction?
That short planning step makes the generation easier to control.
Write a prompt with subject, context, and style
The fastest way to improve results is to make the prompt specific. Instead of asking for "a product photo", describe the object, environment, lighting, crop, and intended channel.
Example:
A clean skincare product photo on a light marble counter, soft daylight from the left, realistic reflections, premium ecommerce style, square crop, minimal background.
That prompt gives the model enough direction to produce something useful for a product card or ad variation. If you are creating marketing images, include details such as the target platform, visual mood, and whether the image needs empty space for copy.
Use reference images when consistency matters
Text prompts are useful when you are exploring. Reference images become important when the output should follow a real product, a visual layout, a character, or an existing design direction.
Use a reference image when you want to:
- Keep a product recognizable across multiple images
- Create a new scene around an existing object
- Match a preferred composition or camera angle
- Explore variations without losing the original visual direction
For brand or product work, reference input usually gives better continuity than prompt-only generation.
Choose an aspect ratio before generating
Aspect ratio affects how useful the result will be. A square crop works well for ecommerce thumbnails and many social posts. A portrait format is better for mobile-first ads and vertical feeds. A landscape or wide format can fit website hero sections, banners, and presentation visuals.
Choosing the format up front prevents extra cropping later. It also helps the model compose the image around the final destination.
Generate from the workspace
image2img runs generation from your signed-in workspace. That keeps prompts, reference inputs, credit usage, job status, and generated assets tied to your account.
After you submit a job, the image moves through processing states until it is completed or fails. Completed assets stay available in your workspace, so you can review them later, download useful results, or continue editing from a saved output.
Review results like a production asset
When the image is ready, check it against the job it needs to do:
- Is the subject clear at thumbnail size?
- Does the lighting fit the product or campaign?
- Is there enough space for text if the image will be used in an ad or hero section?
- Does the output match the reference closely enough?
- Would the image need another variation before publishing?
For most real workflows, the first useful result becomes the starting point for iteration. Save the strongest version, then adjust the prompt or reference to create a tighter variation.
A practical first workflow
For your first image2img workflow, start with a simple asset that has an obvious business use:
- Choose a product, service, or content topic.
- Write a prompt that includes subject, setting, lighting, and format.
- Add a reference image only if consistency matters.
- Generate inside the workspace.
- Save the best result and create one variation.
This gives you a realistic sense of how image2img fits into product visuals, ads, content production, and website graphics.