This translation has not been editorially reviewed yet. The German version is authoritative. Deutsche Fassung
SEO
How do my images get into Google Image Search?
Google finds images when it reads your pages; You do not have to register them individually. An image must appear as an ordinary image element in the source code of a page that Google is allowed to read and index. Images that are only included as background via the style sheet are not indexed by Google. How and where a found image appears depends on the alt text, the text around the image and the page on which it is written.
Google must find the image in the source
Google reads images from the src attribute of image elements, even within a picture element. Ordinary HTML image elements help the crawler, in Google’s words, to find and process images. The formats BMP, GIF, JPEG, PNG, WebP, SVG and AVIF are supported. If your website delivers different image sizes depending on the screen, Google recommends that you always provide an avoidance address in the src attribute. The same image should be included everywhere under the same address so that Google does not have to retrieve it several times. The robots.txt file must not lock images that should appear.
An Image Sitemap Calls Google Images It Else Overlooks
A sitemap is a list of your site’s addresses for search engines. She can also name their pictures on each page. Google describes the purpose as follows: Report images that it would otherwise not find, such as those that first download a script. Whether you write the image details in the existing sitemap or create your own image sitemap is the same for Google. Up to 1,000 images are possible per page. If the images are on another domain, such as a delivery service (CDN), the sitemap may still name them; Google recommends confirming both domains in Search Console.
The page around the image determines where it appears
According to Google’s documentation, the content and details of the page in which an image is embedded have a major impact on how and where it appears in search results. Google takes the subject of an image from the page content, captions and titles and advises to place images next to matching text. The file name gives only very slight indications; "kochinsel-eiche.jpg" still says more than "IMG 2481.jpg". Also for an image result Google generates title and text excerpt of the page automatically. A separate page title and a suitable description also work there. What a good alt text looks like in the article "What are alt texts for pictures and how to write them correctly?".
The preview image can be suggested
Which image Google shows as a preview to a page, it chooses automatically. You can suggest an image: with the indication og:image, which also previews when sharing, or with primaryImageOfPage in the structured data. Google advises on an image that stands for the page and is high-resolution. A general image such as the logo and extreme aspect ratios are not suitable according to Google.
Author and license can be identified on the picture
For photographers, agencies and everyone who licenses images, there is a separate award. With the type ImageObject you specify the address of the image file and at least one of the details author, image credential, copyright notice or license. Google can then show in the image search who the image comes from and how it can be used. For the marking as licensable, the licence indication is mandatory. The same information can be written into the image file as IPTC data; Both contradict each other, Google takes the structured data.
- Filter the Search Console performance report by the "Image" search type. It shows which pages are displayed and clicked on via the image search.
- Ask your agency if the large images on the homepage are included as a picture element or as a background in the stylesheet. Only the image element can appear in the image search.
- After moving the images to a delivery service, check whether its robots.txt shares the images and whether the sitemap names the new addresses.




