Economists and other people who use “archival data” love to make inferences from data. Archival data is otherwise known as behavioral data, like the kind you get on the web by watching what people do. Inferences are just guesses …. educated and otherwise.
Economists, in fact, don’t see the relevance to speaking with people to find out what’s in their minds because of their theory of “revealed preferences”. That is, why ask someone what they think, just watch their behavior and that will reveal to you what they were thinking.
When I was trained in marketing research this type of approach was considered, well, quite a stretch. For most marketing concepts (except for purchase behavior or cold hard cash), what you’re trying to measure is just a concept or idea. There is no one-to-one relationship between the concept and something you can measure. Instead, people take the hard data and imbue it with meaning.
So you might not find it surprising how perplexed I’ve become over the explosion of available data on the web and the inferences that are currently being made about archival data. The biggest sensation is how data can reveal how much influence and authority you have. Technorati, for example, has an “Authority Index” for measuring a blog’s importance and authority. You can now find several website that will calculate your influence, how “hot” you are, etc., using Twitter data. It’s really gotten out of hand.
I think most of these measures are going way beyond the data into something that borders on wishful thinking. Here’s why.
Most people don’t know this, but as an academic researcher, and one who had to publish to get tenure, the golden rule has not only been getting a lot of publications but also being cited by a lot of other researchers. The key indicator of this is the “citation index” …. or the number of the other people who have cited your research in their research. This citation index is basically the idea behind Google’s rankings. You might think of this as equivalent to getting a lot of follower’s on Twitter.
What academics love to do is imbue this index with meaning, such as a high index means a person’s work is really important, or even innovative or, better yet, path breaking.
The problem with this index is the incentives that call the meaning of the index into question. For example, you might cite somebody else’s research because you found a flaw in it, or because you want to make your research look more scientific by citing someone else, or because you want to support a finding and need to find someone else who said it first (in academics, unlike the real world, you have to make sure what you say hasn’t been said before). The point is that there are several reasons a person might cite the work of someone else that has nothing to do with how innovative, important, or path breaking it is.
So, you can see the problem I have when I hear about an Authority Index, for example. This index is defined as the number of blogs linking to a website within the last six months.
This is the same as a citation index. It’s just a number (number of blogs linking to a website), which is then imbued with meaning (authority). By calling it “Authority”, people will naturally think about how important the blog is, and how pathbreaking and innovative the ideas are. But nobody talks about the incentives that call this meaning into question.
The big deal these days is using the “retweet” and “follow” data in Twitter to measure how important and influential you are. For example, someone is now offering $125,000 in order to lock in a way to gain hundreds of thousands of new followers.
This is just a marketing ploy, of course. Nonetheless, the idea of having thousands of followers has taken on a grandiose meaning, well beyond the simple fact that it’s just a simple number.
In fact, from what I’ve seen lately, I’ve received hundreds of followers who are basically scam artists where they are trying to get you to come to their twitter page where they’ll show you how to make a million bucks on some gimmick. Hmm, are these the followers that will add to a number that other people will assume means my tweets are really valuable?
This problem of archival data and placing ideas on it has plagued marketers for a long time. There are ways of dealing with this, but that’s not the point of this post. My purpose is just to make the statement that before believing what archival data means, we might all just pause a moment and remember that inferring ideas from archival data is a risky business and shouldn’t be taken too seriously.
So, what do you think?
Is having more followers really a measure of how important and influential a person is? Do you use this as a “cue” that informs you about someone? Do you think people who retreet other’s posts really believe they are influential? Do you change your behavior and do marketing in a completely different way (which is really a measure of influence) based on a high follower person’s tweets?
Great points Allen.
You have hit on one of the important misconceptions of social networking for marketers. That is, many think that large numbers of followers equals success. Not true. What counts is, first, the number of qualified followers and, second, how well you engage with them in order to get them to buy something (or do something that advances the sale).
It is not surprising since that is exactly how many feel about advertising in general. In fact, the concept of branding (in my opinion) has a lot to do with the fact that it is often difficult to measure the success of ads. So if you say you are branding it doesn’t mater if you cannot track sales. All that maters is how many supposed impressions you get.
Don’t get me wrong, I am not saying branding is useless, it has value for commodity products such as soft drinks, water and the like. In social networking t helps to be part of the community, but it helps a lot more if you are an active participant and provide value to that community.
I have always had a problem with the notion of authority being tied to number of Twitter followers. Just as there are ways to manipulate impact factors, there are plenty of ways to cultivate thousands of Twitter followers.
Charlene Li added some confusion to the whole numbers game issue when she listed, as one of her 2009 Social Media predictions, “Having thousands of friends becomes “so 2008″ and defriending becomes the hot new trend.” If she were to be right-which I personally can’t imagine because of the numbers/authority concept, then I guess marketers would be seriously out of luck.
Instead of filling up your comment section, here’s a link to a post I wrote about this issue:
http://www.mizzinformation.com/2009/02/which-is-it-quantity-or-quality.html
The problem is that we have a few folks (not all) who are called experts within the social media world, who have little to know real-time experience in a business, their’s or someone else’s. Their experience comes from the Web 2.0 environment and they put their own spins on things such as ROI, Value, Authority, Influence, and so on. These same folks sometimes say things that are not supported by the research or by understanding how to measure business success.
Allen,
This is a great post. Thank you for saying what I have been thinking. I spend a lot of time simply shaking my head in wonder at the value being placed on quantity versus quality. My definition of an authority is someone who knows how to do something and has demonstrated the ability to either do or lead others to accomplish it.
Today’s social media world is in its infancy. It will evolve. When it does, the people who are measuring their expertise by followers, links, and retweets will have to prove their value or they will disappear. It is similar to the early days of the Internet. There were many “experts” who didn’t know much more than the corporate teams hiring them to market new websites. If you asked them to explain how everything worked together for corporate growth, the common answer was “It’s complicated.” In other words, you weren’t smart enough to understand.
Everyone who is a true authority has one thing in common. They can (and usually will) explain the process and value to anybody.
Social media is a valuable tool for businesses who want to build relationships with their customers. Before hiring experts to lead your company, do some legwork yourself. You may find that you are the authority.
TT, nice one. On the other hand, I’m just wondering if there isn’t any simmilarity to political voting?
When a politican is popular (gets many votes), it might not be for him being smart or the best leader. It might just be him having a lot of money to get these votes? And all we’re measuring to make him a leader, is the number of votes.
As for me, I find this funny – those with some less “authority index” are always more interesting. People that are No1 are mostly just trying to stay there.
Hi Allen and thanks for the post.
One thought to tack on to the relevant comments above: deriving “authority” based on link analysis only misses some of the point of the content itself. A more refined approach toward estimating “authority” would include categorizing the content. Meaning, the “popularity” (however defined) of a blogger can be refined by taking into account the primary focus of the material.
For example, if most of your tweets are scams and you have tons of followers, your influence ranking would be high for “Scams” but low for “IT Security”. Understanding the context helps us modify our position on “authority”, IMHO.
Not sure how clear this is, time for a cup of coffee. Your points remain valid, I just don’t want to throw out the baby with the bathwater!
Dean
Hi Allen,
Your “question authority” graphic says it all. It is really unfortunate that so many people are obsessed with a false barometer of popularity. I wonder if the people who seem to collect followers as a badge of self-worth have even a tangential relationship to more than a handful of the people in their networks.
I’m not sure why we even use the word “authority” in these contexts. I’m much more interested in the concept of influence or reach, which I tend to think of in the context of the value you bring to your network and the value your network brings to you.
Great post; I really like the academic research analogies.
Best,
Daria
Hi All
Thanks for your posts. Everyday I’m seeing more of these authority and indexes pop up, so I’m glad I’m not alone thinking a lot of this is just fun and games and shouldn’t be taken too seriously.
Allen
Like measures of social network analysis, all of these “authority” and “influence” metrics really can only be indicators of potential. Somebody with few followers/subscribers/linked blogs probably can’t seed a viral conversation like somebody with tens of thousands… but tens of thousands is no guarantee either. If there was a way to capture packets flowing *through* social networks, it would be another matter to define who has, and does not have, actual authority, influence, and power.
[...] Allen Weiss questions the definition of ‘authority’ in The Meaning of Numbers: Does an ‘Authority Index Equal Authority?. [...]