What You Need To Know About Klout, Alexa, and Other Social Scoring
In this low-attention-span digital world that we live in, we need any shortcut we can get. This is never truer than in the Influencer Marketing industry with social scoring. There are many different shortcuts we can use to see how effective someone’s social presence is, how much blog traffic they get, and how “influential” they might just be on social media.
Here are some of the social scores you might look at before approaching influencers for an outreach campaign:
1. Klout, Kred, and PeerIndex Scores
First up, you have “influence” scores – Klout, Kred, and PeerIndex. Klout is the clear leader in this category, stating that they are the “standard of influence.” I personally think they did more damage to themselves with that statement than simply calling it an engagement score. Does it actually measure how influential someone is? Depends on what you’re trying to measure – if a re-tweet or a like is considered influential to you, then yes. If a conversion, a trial, purchase, change of opinion, etc., is what you consider influential, then NO, Klout does not measure that.
These three scores look at various social media activities, such as likes, re-tweets, mentions, replies, shares, comments, etc., and try to estimate how influential you are on social media. A problem with this type of general social engagement score is that it doesn’t distinguish very well between the networks. You might be really good at spreading the word on Twitter, but suck at Facebook. If you want to do a Facebook campaign, and want influencers on Facebook, these scores might be really hard for you to use. In the future, it would be better if each social network would have their individual scores, plus a general social media score.
Another major flaw with these scores is that they can easily be gamed. People artificially inflate their social scores to make themselves look more influential than they are. Although a lot of people game the score, you could argue that only the “elite” influencers actually care about gaming their score, and not the actual mass social media users. But that small percentage of users that do game these scores kind of taint the rest of the industry. You can use these scores as an initial sorting mechanism, but deeper research needs to be done on your part.
2. Alexa Rank
Alexa tries to estimate how much traffic a website gets, and then gives the website a ranking. They do this by measuring clicks, traffic, duration, etc., from people that installed the Alexa Toolbar on their browser, and then do some calculation to estimate real traffic. Sometimes it’s accurate, and sometimes it isn’t. Alexa also looks at website reputation by measuring how many websites link to yours.
I’ve often seen some blog traffic that Alexa either overestimates, or underestimates. If a blogger registers with Alexa, they will have verified metrics that will show the actual number of unique visitors per month that the website gets. Most people don’t verify their site though.
Always ask the blogger for an estimate of how much traffic they get per month to their site, and see if it matches the Alexa rank. You could also ask for a Google Analytics report from the blogger, but some might be turned off by this.
3. Other Social Scoring
There are all sorts of different social scores you can look at that aren’t the “big 3.” If you have a PeekAnalytics account, you can get a social pull score which measures how influential your Twitter audience is across 60+ social networks compared to the average Twitter user (this is a measure of the AUDIENCE, not the INDIVIDUAL – much better metric).
Radian6 (now Salesforce Marketing Cloud) has their own magic formula to determine influence. What’s nice about their score though is that you can manipulate your “influence” criteria based on what’s most important to you. If the number of comments and twitter followers are important to you, set that as your top weight. If Klout is important, you can set it for that. It then looks at keywords in your topic profile, and tries to give you a list of influencers that are most active in that topic, including celebrities, brand accounts, publishers, and long-tail influencers (the really important ones).
And then there are other websites trying to give you some sort of measure of influence. LinkedIn recently joined the game with endorsements. Connect.me looks at online reputation, and tries to give you a “trust” score, where you “vouch” for someone’s skills/topics instead of +K or +Kred.
LittleBird gives you an insider score, trying to measure who influences the influencers.
And there are several other scoring algorithms, too many to list in a single blog post.
What’s missing from these scores?
There is one major flaw with social scoring though – you can’t measure quality. That’s one thing that cannot be automated… how do you automate the judgement of quality web design? Quality of conversations? Quality of blog comments? Quality of writing?
You can’t. That part will always have to be done by a human being. Our InNetwork Score has a manually imputed (by the Influencer Relationship Manager) quality score that looks at quality of writing, quality of web design, quality of engagement, etc.. There’s no other way of doing it. There’s also a Marketer Rating involved in the score, where a marketer rates the performance of an influencer post-campaign. The “human algorithm” is something that’s really important, but often missing from current social scoring.
The bottom line is – if you take a look at an engagement score, go deeper to find out what you can about the quality of the influencer. There are things that an algorithm simply can’t measure. A social score is a good starting point to filter out people, but don’t leave it at that – go further in your analysis.
What’s your opinion on social scoring? Please leave a comment below!