2

I have 4 sizes for a single image in a page of my eCommerce website. 600x600px , 350x350px , 220x220px , 110x110px

There are 3 solutions:

1- Loading the big image (600x600px) from server and cache it, then generating thumbnails using the cached one by a client-side plugin.

2- Loading the big image and thumbnails all from server. (in this case, thumbnails are generated in server)

3- Loading the big image and create thumbnails by resizing the big one using CSS. (or for example we can load 600x600px and 350x350px ones and create thumbnails by css from 350x350px one)

Which solution is the best for SEO ? or if there is any other way, I appreciate.

Improve this question
3
  • A 600x600px image isn't very big for a mobile, why not just use that? – Simon Hayter Jun 24 '15 at 5:30
  • actually i am using that, but the matter is the SEO for thumbnails .. – Payam Shakibafar Jun 24 '15 at 5:43
  • You should definetely go for thumbnails scaling images with CSS is one of the most common SEO mistakes. It's pretty simple you are loading 600x600 image once which might take browser 0.2 ms but the problem is that you make second request which can take the same amount for the browser to resize the image and calculate the proportions, it results critical on your SEO so use thumbs. – knif3r Feb 24 '16 at 16:30
1

Solutions 1 and 3 are pretty much the same, with #1 taking an unnecessary step for a negligible saving on bandwidth - as the CSS resizing at #3 is fine enough. For really saving bandwidth, with the implied SEO benefit that the site would load much faster, go with solution #2.

Improve this answer
0

The efficiency of bandwidth seems to be factored into SEO though only the search engines will know the secret weighting they apply to it.

If for example your website uses a lot of images that are downloaded at higher resolution than is required and then they are scaled down to a fixed size in the HTML/CSS this can receive an SEO penalty as they think you should supply the image at the right size your page uses to reduce bandwidth use and client-side processing.

However, for circumstances where the thumbnail size required depends on the client device screen size or orientation, CSS media queries can be used to specify the right thumbnail image file as appropriate and then apply scaling using CSS in places where required. The SEO penalty for images of the wrong size does not seem to apply to images specified as part of CSS media queries.

<style type="text/css">
.thumb {
  background-image: url('https://lh4.ggpht.com/wKrDLLmmxjfRG2-E-k5L5BUuHWpCOe4lWRF7oVs1Gzdn5e5yvr8fj-ORTlBF43U47yI=w300');
  background-repeat: no-repeat;
}
@media (min-width: 600px) {
  .thumb {
    background-size: 600px;
    height: 600px;
    width: 600px;
  }
}
@media (max-width: 599px) {
  .thumb {
    background-size: 300px;
    height: 300px;
    width: 300px;
  }
}
</style>

<div class="thumb"></div>

An image can be scaled in CSS, for example if you had a 300px by 300px image, you would use max-width/height to scale down, min-width/height to scale up (with some images this will not necessarily look terrible):

<style type="text/css">
img.thumb110 { max-width: 110px; max-height: 110px; }
img.thumb220 { max-width: 220px; max-height: 220px; }
img.thumb350 { min-width: 350px; min-height: 350px; }
img.thumb600 { min-width: 600px; min-height: 600px; }
</style>

<img class="thumb110" src="example.png" />
<img class="thumb220" src="example.png" />
<img class="thumb350" src="example.png" />
<img class="thumb600" src="example.png" />

The best method will depend on how your website works and which will in the end mean the least bandwidth transfer in downloading content required for the page.

If you're getting the server to do the heavy lifting in terms of generating resized thumbnails on demand, then at least for the ones generated, save them for future use to save your server having to keep doing this and improve the page speed.

For any e-commerce pages that show alot of thumbnails you may be better embedding the thumbnail images as data within the HTML rather than having the browser perform a separate HTTP GET request for each image. For example:

<img src="data:image/jpeg;base64,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" />
Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.