High resolution (retina) images vs load time

[deelo555]

I have an ecommerce website and have a product slider with 3 images.

Currently, I serve them at the native size when viewed on a desktop browser (374x374).

I would like to serve them using retina image quality (748px).

However how will this affect my ranking due to load time?

Does Google take into account image load times even though these are done asynchronously? Also as its a slider, its only the first image which needs to load. Do the other images contribute at all to the page load time?

[deelo555]

"Large pictures tend to be bad for user experience."

I disagree. I think what you mean is slower loading is bad for the user experience. Higher quality pictures are better for the user experience.

I've been looking into deferring loading of the additional slider images. That should definitely improve load time as all the bandwidth can be used to download the first slider image.

Also the first slider image if you use a progressive format should show something quickly and then improve over time.

[mathamatix]

You also have to keep in mind that users will access your site from mobile devices and that the larger the page the longer it takes to load fully. You may lose some people during the time it takes to load the page. My website used to have a slider with three images. i removed the slider and replaced it with one static image. Large pictures tend to be bad for user experience.

[Marcus_Miller]

Hey Dwayne

They are big images but from experience I have never seen a meaningful impact from these kind of changes (in around 15 years). Maybe work on optimising the images themselves as best as possible to bring the overall size down as much as possible. Sure, if your site is a slow loading nightmare and this is just the final straw then it may be an issue but by the sounds of it you are already taking that into consideration and your site is well hosted and performs better than most of everything else out there.

But, as ever in this game, my advice would be to be aware of possible implications, weigh up the pros and cons and then test extensively. If you see an impact in your loading time and search results (and more importantly in user interaction, bounce etc) after changing this one factor then you know you can roll it back.

Hope that helps

Marcus

[deelo555]

Hi,

Its not that small a change...the size of each image will quadruple from around 10kb to 40kb. As there are three images thats 90kb more data. Which is around 20% of the total page size.

That's interesting what you mention about the first byte load time. I would have thought that was overly simple and would definitely have assumed Google would actually be more concerned with how long it takes for the page "to load" (e.g. using their pagespeed metrics).

I've optimized my site extensively and have pagespeed score of 95% and I use the amazon AWS servers.

I agree with your idea about doing what's right for my users. But if Google includes the image load time then my site will rank poorly and then I won't have any users!

In summary, I think what this question really comes down to is how does Google calculate page load times and does this include image load time and does it include load time for all images (even ones which aren't being rendered in the slider).

Thanks,

Dwayne

[Marcus_Miller]

Hey

I think this is such a small issue overall that you should not worry about a slight increase in image sizes damaging your SEO (assuming everything else is in place).

I would ask myself the questions:

• Is this better for my site users?
• does the seriously impact load times (and therefore usability / user experience)?

If you believe it creates a better experience and does not impact loading times in a meaningful way then go for it and don't worry about a likely negligible impact on loading times.

A few things I would do:

• test average loading times with a tool like pingdom: http://tools.pingdom.com/fpt/
• replace your images and test again
• look at other areas where you can speed up loading times
• make sure your hosting does not suck

For reference there was a post here a while back re the whole loading times / SEO angle that determined it was time to first byte (response time) rather than total loading time that had the impact - this would make total loading time academic from a pure SEO perspective but... it's really not about SEO, it's about your site users and whether this makes things better (improved images) or worse (slow loading) for them.

Seriously - don't worry about this small change too much from an SEO perspective. Use it as an excuse to improve loading time as that is a good exercise for lots of reasons but go with what is right for your users.

Hope that helps
Marcus

Ref
http://moz.com/blog/how-website-speed-actually-impacts-search-ranking

http://moz.com/blog/improving-search-rank-by-optimizing-your-time-to-first-byte