Naturally, we’ve been watching the development of Google’s real-time search with interest. Everybody’s probably familiar by now with the scrolling box they’re inserting into search results:
I was reading a bit about what people are speculating is behind the real-time search algorithm and what Google decides to show. An article on eConsultancy got me raging.
The author suggests three things that may be factors in getting Tweets to show in real time search results that I can’t imagine are true – because if they are, they demonstrate that Google has no clear understanding of social media and they contradict everything I tell clients when providing training on how to use social media correctly and effectively.
1. Keywords in usernames = more relevant? Relevant to what?
First, he writes, “Username. Like domain names, keywords may help. Note that Twitter allows you to choose a username (unique) and a real name, so there’s an anchor text thing going on here.”
Now, correct me here if I’m misguided, but social media accounts should be registered to either people or organizations – and it should be clear which is which.
For example, you may be “Paul” trying out Milwaukee SEO and you may be tweeting on behalf of “Sofas and Stuff” but if you call your account “Sofa Sofa Sofa” and you follow me on Twitter, I swear I will block you straight away because that just makes you sound like a spammer trying to get followers to game the algorithm. You pollute the timeline of the site and clog it up with garbage. If, however, you actually call yourself “Paul from Sofas & Stuff” I may consider following you back – if I happen to need a sofa and depending what I find when I visit your page.
If Google is using keywords in usernames as a factor then we’re about to get a load of people registering names and descriptions that read “Sofa Sofa Sofa.”
Furthermore, a good social media account isn’t about the name of the account – it’s about the content. On a website, if you call yourself “Puppy Food, Inc.” it’s pretty likely that you won’t ever have relevant content about sofas so keywords in the name can be an indication.
If you call your Twitter account “Bob Smith”- or even if you call it “LeedsFan67” – you may at some point send out tweets about sofas which may actually be relevant and useful in response to a news story or piece of content relating to sofas.
For example, if a sofa explodes on Big Brother next week then Bob may be tweeting about it. Bob may even be a paramedic and able to discuss the dangers of exploding sofas in relation to fire or offer advice to victims on how to treat their burns.
Clearly, keywords in usernames isn’t an indicator of relevance. In fact, they may be an indication of spam – someone who calls themselves “Sofa Expert” on Twitter, at least in my experience, is less likely to be responsive to people and to engage with the community, or do more than post links to his own spam web content, than someone named “Bob” who’s excited about buying a sofa.
If Google really is using keywords in usernames to determine the relevance of specific Tweets then they’re going to find real-time search besieged with garbage and Twitter is going to have to step up its efforts to weed out spam because it’s about to be even more horribly abused than currently.
2. “Hashtags = Bad.” Or, should that be, “Hashtags=Relevance?”
“One warning sign that Google looks at is whether or not a tweet includes a hashtag. Trending topics on Twitter – largely measured by hashtag popularity – can attract lots of junk tweets, so Google has adjusted its ranking factors for tweets that contain hashtags.
It most likely applies some kind of negative weighting for tweets that contain hashtags, especially those that trend and hit the Twitter homepage (and all user accounts).”
Hashtags are there to identify posts. For example, if you’re at a conference and tweeting about the conference you tag your posts with the relevant hashtag so even non-followers can see them. If you’re responding to a particular story, you tag your tweet with the relevant hashtag. Hashtags represent relevance when used by real users – rather than auto-accounts or spammers.
In fact, when hashtags are abused horribly (Google “Habitat Twitter” if you need an example”), the community reacts swiftly and angrily and a huge negative publicity machine wallpapers the internet with angry messages of derision.
Surely hashtags, combined with other relevant words in a tweet, should be a marker of QUALITY, not the reverse? If Google has actually worked out ways of identifying spam accounts and sockpuppets then surely hashtags can help them understand the relevance and semantically related words in tweets?
3. Use your keywords a lot. But, is Twitter really about keywords?
Finally, the suggestion which really got me fuming is this:
“Keyword focused accounts. If all of your tweets refer to ‘sofas’ isn’t it possible that you’ll be seen as a sofa expert by search engines, and positioned accordingly? Staying on-topic may help.”
So, set up loads of accounts, each based around one specific keyword or set of keywords, use your keywords repeatedly, all the time, in every tweet, every, in the whole world and that will make you an expert – not a dirty spammer? Erm, it makes no sense in a social setting.
If you’re using Twitter well, interacting with people, then you aren’t going to stay on topic. If you do stick to one topic, say, sofas, then you’re not likely to ever really have many followers, are you?
Sure, there are exceptions to the rule like someone tweeting the weather report or football results or something. But if someone is buying a sofa they might follow you for a little while, but once they’ve done making their purchase, are they really going to care anymore?
The way to make them care is by providing added value – talking about more than just that sofa purchases. Or, shock-horror, by getting to know your followers.
Frankly, this one is so open to abuse – and so against the whole point of real-time search – which is to provide relevant, topical, time-sensitive information – that I can’t imagine it could possibly work as a factor. Plus, you’d miss out on the guy in Haiti tweeting about the earthquake as it happens because you’d be too busy looking for tweets from earthquake “experts”- erm people who use “earthquake” in every tweet, that is – and that would actually make your real-time results LESS relevant.
It was all summed up in a very pithy manner by someone on Twitter, as I was writing this:
“Twitter is all about what is happening RIGHT NOW.”
He’s right. Twitter isn’t so much about expertise as about what people find relevant and important at a given moment.
Social media is NOT SEO. Don’t confuse the two.
Let me make this clear – Search Engine Optimisation and Social Media are two different things. Social Media mentions may add weight to SEO efforts and help verify the quality of incoming links, but the way Social Media operates and the way relevance and quality should be established within social media is a very different beast to SEO.
SEO is about explaining the subject matter of pages of a website and website hierarchies to a machine. It’s about helping machines to understand the most important words associated with a given piece of content and ensuring that those important words also match what users type into search engines.
Social media is about people. It’s about interaction, relationships, and communication. It is not about keywords. It is not about links. It is certainly not about how machines understand pages. So if Google is using SEO factors to determine the quality of real-time search then I’m guessing relevance is going to suffer – if not now, then certainly once SEO gets their hands into it.