This week was low on big news, and high on small tiny interesting tidbits, just how I personally like it most. The reason I like that most is that those tiny tidbits usually show you more about how Google and / or other search engines work than the big flashy news items. Let’s dive right in!
An AMP ranking boost
Google is promoting AMP (Accelerated Mobile Pages). We’ve written about it before. I’m obviously not a fan. Google is pushing it hard though, and in their latest release they’ve said that they’ll include AMP pages in the search results as of January and they “might” give them a ranking boost. This annoys me mostly because Google uses the phrase “might give a ranking boost” so often now that it’s no longer even worth listening to.
If everything Google said got a ranking boost did get a ranking boost, spammers would rejoice and big companies would lose all of their rankings because it’s impossible to keep up. So it’s not actually true. It just pushes us to do stuff. In the case of HTTPS / SSL, I’m all in favor and have been telling people why for years.
In the case of AMP I think that’s a stupid idea and we shouldn’t do it.
Google wants your appointments
Google is testing a feature that allows people to make appointments straight from the search results. Whether it’s for your practice, hotel, really whatever it is you have that requires appointments, this might be interesting. They’re testing it but they (accidentally I think) published the schema for it and then removed it, but of course, they were saved. It looks doable. If they go about opening this up to everyone we’ll probably add this to our Local SEO plugin.
Google on 301 redirects
If you create a 301 redirect, you have to really think about doing that. Because in HTTP, a 301 stands for permanent. Which is exactly how Google treats it, after a while. It can take a couple of weeks or sometimes even longer for a 301 to kick in and really transfer the value from one page to another. Once it has though, removing the 301 and getting the old URL to rank again is tough. This is a story about what you could do when you do decide to remove a 301. It basically says: really embed the page in your site, and have patience. Months of patience.
Disavow away
If you’ve been hit by Google Penguin, one of the tools you need to get out of that Penguin penalty is the disavow file. This let’s you tell Google “don’t count these links to my site”, basically saying that those links where bought or are otherwise low quality. The problem is: you can get too drastic in removing links.
This story is typical in that someone got an enormous boost in traffic when he accidentally removed a site’s disavow file. You really should only disavow the bare minimum, and not just disavow tons of our links because you will hurt your site too much. Play with this at your own risk though, the rewards can be big but the consequences can be too.
Mo’ money, less problems
Mo’ money, less problems, at least, that’s what Google seems to think. They’re testing out showing 4 instead of the regular 3 ads above the organic search results. Of course, this doesn’t make SEOs around the world happy. Not that there’s much we can do about it, but we’d rather have Google show our nicely optimized sites in the organic results.
The question is though: is this really surprising? There are keywords where I see people pay 10s of dollars per click. If Google can make sure you almost have to click an ad, it makes more money. That’s what it’s in the world for, despite what its fancy mission statement says.
Google is definitely not in the world to make a social network. It’s now dropping Google+ counts from its ads, just one more sign that the social network is truly dying.
Market share numbers
Every once in a while, Comscore and a few other companies will come out with market share numbers. In Europe, we’re all agreed that Google has like 95% of the market and usually more. In the US, that marketshare number is always remarkably lower. Remarkably, because most of the sites I’ve worked on in the US get 90-95% of their organic search traffic from Google as well, whereas it “should be” 70% or so according to those marketshare numbers.
This piece of research shows some signs as to those numbers actually not being very factual. At the same time, one of the reasons we’re seeing the difference between reported marketshare and actual traffic to websites is because Yahoo! and Bing are very effective at sending traffic to sites in their own portfolio.
Machine learning, human doing
If you’re interested in how Google uses machine learning, read this post. Basically John Mueller is quoted in it saying they have to balance machine learning and human decisions, which I’m guessing is hard. This part of the quote is key:
Sometimes it makes it hard for us to debug and diagnose what is actually happening there.
Googles sometimes don’t know anymore why something ranks where it ranks. I’ve noticed that before. It’s funny but not all that surprising given how many changes go in to the algorithm, the machine learning only makes it even harder.
That’s it, see you next week!
