Eventually, One Of These Updates Will Make Google News Not Suck At Tech News, Right?
![](data:image/svg+xml;charset=utf-8,<svg height="152" width="211" xmlns="http://www.w3.org/2000/svg" version="1.1"/>)
![](data:image/svg+xml;charset=utf-8,<svg height="425" width="620" xmlns="http://www.w3.org/2000/svg" version="1.1"/>)
For as long as I can remember, there has been one constant in tech news: Google News sucks at it.
Now, that’s not entirely fair since Google News doesn’t actually provide any of their own content. Instead, they use the supposedly magical Google algorithms to curate others’ content from around the web. Still, there’s just no way around it — the product, at least for tech news, sucks.
Again, this is nothing new. I’ve been writing this article at least once a year for five years or so. But what’s amazing to me is how many updates Google does to the product and it still sucks. Even more remarkable is that a lot of people continue to go to it as a source of tech news. How do I know? When TechCrunch does appear on it, it sends a lot of traffic.
So is this just me complaining that we don’t appear on it enough? Well sure, partially. But to me, our sporadic appearances are also indicative of the big problem the service has.
How much tech news gets broken on TechCrunch? Even our rivals would admit “a lot”. Yet if that news appears on Google News at all (far from a given), it often does hours later and almost always under another source that has re-reported it after we have. That’s really Google News’ problem in a nutshell. It’s hit-or-miss, way too slow, and as a result, often credits the re-bloggers rather than actual sources.
Let’s look at the Google News Technology section right now (which, humorously, is still called the “Sci/Tech” section — Technology apparently isn’t yet big enough to have its own stand-alone section even though I think just about everyone in the media industry would disagree at this point — and impressively odd since Google is a company known for one thing: technology).
- Top story: Sony’s freebies after Playstation hack. First of all, a fairly “meh” top story. (Main story is very old, new update today is fairly lame freebies in response.) Second and more importantly, even this new twist to the story is hours old. They use this Computerworld story as the main headline, which is little more than a repurposing of information from Sony’s own blog from hours earlier.
- Second story: RIM stock falls on Playbook recall news. Again, old news — from yesterday, in fact. And they link to this Reuters story which appears to have been written on a typewriter in the 1940s — and does nothing to expand upon the story which is over a day old. The secondary headline they choose is from a site called “Today’s THV” which is some local bullshit site that just republishes AP stories. Puke.
- Third story: “Next hurdle for Giffords: Mending damaged skull.” Interesting and important story, no doubt. But a top tech headline? The algorithm is fooled because her husband is an astronaut aboard Space Shuttle Endavor, which took off on its last flight earlier today. Again, that would also clearly be a “Sci” story, which should have its own category.
This goes on and on.
Now, let’s compare it to Techmeme.
- Top story: Rumors of Apple secret launch for 10th anniversary of Apple Stores. Hot story? Yep. Bullshit? Maybe, but judging from the smoke surrounding this, probably not. Regardless, it’s something everyone wants to know about. And yet, it’s story number nine on Google News, waaaaay below the fold. And they cite PC Magazine which is re-blogging the actual source, BGR. Guess who Techmeme cites? Yep.
- Second story: Bing/Facebook integration. Again, good story, though several hours old — BUT, Techmeme cites the Bing blog as the big source of the news. In other words, they give proper credit rather than award johnny-come-lately, which is exactly who Google News gives credit to in their listing for it (more on this in a minute).
- Third story: Amazon rumored dual-core and quad-core tablets. Again, really hot story. And at least at a high level, seemingly for sure not bullshit. Google News doesn’t have this story at all. Not one word, even though a dozen or so big sites are covering it.
Techmeme learned long ago that algorithms alone aren’t nearly enough to fully, quickly, and competently convey the news. Google has not yet learned this. For some categories, their algorithms probably work fine. For tech news, they suck.
And I’d argue they suck for both visitors and for publishers. Because it’s so obvious what Google’s algorithms are looking for, there’s a not-so-secret trick on how to game them. Instead of rushing to get a post out about a hot story, just wait a few hours. By then, the story will bubble towards the top of Google News’ tech section and if you time it correctly and you’re a site that Google News watches (some good, some bad), you can easily be the top headline for the entire section (which also means placement on the main site as well).
This works because Google News favors recent “takes” on a story instead of actual sources of information. You could argue that’s okay in some situations, but often these “takes” are just like the Today’s THV bullshit: republishing AP content late.
In other words, if you’re in the business of breaking news, you’ll almost never find your stories on Google News. Instead, you’ll find someone else re-purposing your story there hours later and reaping tens of thousands of pageviews as a result of the sloth and/or jackassery.
Why bring this all up now? Because Google announced yet more updates to Google News today. I tried reading over what’s new, but my eyes quickly glazed over. I’m certain it will be business as usual.
Actually, it might be worse.
As Danny Sullivan points out, Google now allows you to cut out the source of much of the fresh content — blogs — as an option. Brilliant. Now you can further hide the sources that Google was already burying for you.
At some point, I have to believe Google will start to care about the quality of this product. But years of experience tells me otherwise. And really, that’s fine by me. It’s just another service I can easily ignore.
![](data:image/svg+xml;charset=utf-8,<svg height="125" width="187" xmlns="http://www.w3.org/2000/svg" version="1.1"/>)
![](data:image/svg+xml;charset=utf-8,<svg height="105" width="187" xmlns="http://www.w3.org/2000/svg" version="1.1"/>)
![](data:image/svg+xml;charset=utf-8,<svg height="116" width="187" xmlns="http://www.w3.org/2000/svg" version="1.1"/>)
![](data:image/svg+xml;charset=utf-8,<svg height="135" width="187" xmlns="http://www.w3.org/2000/svg" version="1.1"/>)
