Hi guys, sorry if this has already been discussed but couldn’t find anything.
Our site thecarexpert.co.uk is registered on Google News and we are starting to draw reasonable traffic from it. However, our results always showed up with either no image or someone else’s image attached to our story.
We tracked the issue to the CDN – when images where routed via a CDN (we have used KeyCDN and Jetpack’s Photon), the images would not display. When we switch off the CDN, the images display correctly in Google News.
I would obviously like to run all of our images via CDN for site performance, but not having a thumbnail hurts our Google News rankings and clickthroughs.
Can you suggest where I should look to try and sort this?
Many thanks,
stuart
Hi,
Unfortunately i can’t say for sure why this happens. We didn’t do extensive tests with Google News yet, but we have it on our list
Google guideline for images can be found here – https://support.google.com/news/publisher/answer/13369
If everything is right then the issue must come from the way google interacts with the images delivered by CDN.
Use the browser inspector to check the page source, look for the og:image property to see if it changes once the CDN is active. You can also make some tests with cloudflare, see if it behaves the same.
Thank you!
Thanks Emil. I have also been running WP Rocket, which I am currently swapping over to WP Super Cache and Cloudflare to see if that improves performance, so hopefully I will get it sorted eventually!
Just as an update to this for anyone else having similar issues, we are now running WP Super Cache and Cloudflare and Google News thumbnails are working correctly!
We went through this in 2013 when trying to use Amazon S3, and even with a sub-domain for images setup as cdn.oursite going to the image bucket, Google didn’t quite know what to do with stuff. Fact is they like to still ‘scrape’ the image from the page, even if it’s effectively in your news sitemap XML file properly. You may also find, putting two images in a post, specifying one as primary that shows in OG, twitter and XML as “the image,” they may scrape different one on the post. Kinda stupid there.
Fact is they still use fall back “content scraper” in tandem with reading the news map XML file you put into your search console.
Great that you got it sorted! 🙂
