Skip to content
Get started

Design feedback from a screenshot

Upload a screenshot of a web page or an app, and Do4U gives you numbered findings: buttons that are too small to tap, text that is hard to read, words that are part of a picture, a hierarchy that hides what matters, and uneven spacing. Each finding has a severity, what the AI sees, one concrete change and a frame on the image that shows where it is. This is an AI assessment, not a measurement. The AI cannot measure pixels or contrast, so a pixel size, contrast ratio, percentage or colour code that it claims is taken out of the text and listed. Where an area is also given in pixels, it is calculated from the rough area and marked as an estimate. Your image is shown only in your browser and is never sent back.

Try it now Pricing

Who it is for

For designers, shop owners, agencies and developers who want a quick second opinion on a screen before they build or publish it, and for teams that want findings they can copy into a ticket.

Typical uses

  • Check a design draft or a live page for buttons that are too small to tap
  • Find text inside images that should be real text
  • Get a second opinion on a screen before a launch
  • Copy the findings as a Markdown list into a ticket

Example

Input

Screenshot of a shop page on a phone. Note: online shop for garden tools, mostly used by older people on their phones.

Result

1. Target size (high): the cart button is much smaller than a fingertip. Suggestion: make it at least 44 px tall. Frame: lower right. 2. Contrast (medium): the light grey price on white is hard to read. Suggestion: use a clearly darker grey. 3. Text in image (low): the delivery promise is part of a banner picture.

A fixed example, shortened. Real results are AI-generated: please check them.

Price and limits

Price
usually 2–3 credits
Available with
Free, With package
Input limit
One page with up to 600 characters of HTML

Privacy

Processed on our own servers in the EU (Poland). Inputs and results are deleted after 7 days (Free) or 30 days (packages), or right after you see the result if you choose not to store them. Images are deleted right after processing.

In the API

Tool name: design_feedback. Create a job with POST /api/v1/jobs and fetch the result with GET /api/v1/jobs/{id}. You find your API key in your account.

curl -X POST https://do4u.eu/api/v1/jobs \
  -H "Authorization: Bearer d4u_…" -H "Content-Type: application/json" \
  -d '{"tool": "design_feedback", "url": "https://example.com/"}'

Full API documentation →

Questions and answers

Does the AI measure pixels or contrast?

No. The AI sees the picture at a fixed internal resolution and gives an assessment, not a measurement. If it claims a pixel size, a contrast ratio, a percentage or a colour code, the code takes it out of the text and lists what was removed. To measure contrast on your real page, use the accessibility check.

How do the numbered frames work?

Each finding comes with a rectangle in percent of the image. The code checks that it lies inside the image and is not empty; otherwise the frame is dropped and the finding stays. Select a finding or a frame, with a click or the Enter key, to highlight the other one. Your image is shown only in your browser, from the file you chose, and is never sent back from our servers.

Which kinds of findings do I get?

Target size of buttons and links, contrast, text inside images, hierarchy, spacing, readability, consistency and other points. Similar findings for the same area are merged, the list is sorted by severity and limited to a maximum number, and you can copy it as text or as a Markdown list.

Is my text stored?

Only as long as needed: 7 days on Free, 30 days with a package – or not at all if you choose not to store it. Processing happens on our own servers in the EU (Poland).

Can I use the result right away?

Results are AI-generated. Please read them before you publish or send them.

Related tools

  • Accessibility check

    Checks a page for common barriers and explains how to fix them.

  • Alt texts

    Short image descriptions for screen readers and search engines.

  • Inclusive and plain language

    Points out wording that excludes people or is needlessly complicated, with neutral or simpler alternatives. Suggestions, not rules; whether to use gender-inclusive forms is up to you.

All tools →