Measuring Influence vs. Popularity

I had an interesting talk with some folk on Twitter this morning about not mistaking popularity with influence. In fact, I find this to be among the most misunderstood issues in measuring social media. Popularity is easy to measure. All you do, is look at the total number of visitors and/or the frequency of visits. You can increase sheer numbers by a few tricks the the SEO crowd will tell you about, or by posting frequently or by extending a conversation cited on TechMeme. You need to be frequent and you need to drop a lot of big names names like "Google," "Apple," or "Obama."

A killer headline for the SEO folks would be "Googling Apple for Obama followers who use Firefox." You'd get a ton of traffic, but would it be relevant to your purposes? And would you be influencing anyone at all.

But numerical totals actually tell you extremely little about influence. People could be coming because they hate what you are saying and want to keep an eye on the opposing side. I follow a couple of political blogs, for example, where I fundamentally and passionately disagree with the authors.

For a very long time, the site was called Naked Conversations, which made good sense since it was started as a place to transparently write a book of the same name. It took s a while to discover that the site name was inflating our traffic. People were going to Google and typing in graphical variations of the keyword "Naked" and this site popped up. The more searches, the more prominent we became in certain circles that were decidedly not influential to business-oriented book buyers.

Let's do a reverse example. Suppose I were a political blogger and I had an audience of just three followers. Those followers were very engaged because they read everything I posted. They commented often. They took what I said and quoted me to other people in other conversations. But there were only three of them. Therefore I would be ranked lower than chopped liver in all the ranking systems. The catch is that those three readers were the President of the US, and the heads of China and Russia.

Influence is extremely difficult to measure. First, there are different influencers by different topics. Second, as much as I cheerlead the power of the conversation, most content continues to be read, absorbed, shared and discussed by people using RSS who do not join our conversations where we can see and count them.

Engagement is one way of assuming influence. If people come back every day and stay for long periods we assume it's engagement. It can also be that people just frgot to close a tab after reading you and walking away.

I have a great deal of respect for a score of bloggers and perhaps two score of Tweeters. I refuse to share there names because, the names steadily change over time. Some of the names invariably are people with very few followers, who address matters such as hiking gardening or the environment. I talk little about them because these are topics that do not interest most of the people I am trying to influence on the topic of social media. When I start writing about something new, such as China, I gain some readers and hopefully influence them, but I also lose some who do not care to read what I have to say on that topic.

So, what's my point? I think there are two:

(1) You need to think, really think, about who you read and which of them truly influence you. Then cross check and see if they are the most popular. Smetimes there will be a direct correlation. Other times not, I would guess.
(2) The best way to influence thers is not to try to be popular. It is to shae what you know and care about. Those who care about the same topics will find you. The best conversations do not need to be held with the largest crowds.

Comments

[Phil Baumann]

I think there's another word worth mentioning here: meaningfulness.

Popularity can be an influence and influence can be popular. Influence may appear to be the more 'nobler' of the two but there are many powerful influencers who offer little meaningfulness and there are popular sources of meaningfulness (rare of course).

You make an important distinction between the two words. There's nothing wrong with either, per se but neither gets us to what matters: meaning.

When I make choices as to who I follow (or lead), neither influence nor popularity are as primarily relevant to me as meaningfulness. If it's not meaningful, who cares how popular or influential the source?

[Tom]

Your post reminds me of question/answer challenge I recall from long ago that was posed by a now retired technology PR professional (that might have been you.)

Would you rather have 50 journalists from 50 dailies and weeklies at your product launch press event or just 2? And what if the 2 journalists present were the technology editors from the NY Times and Wall Street Journal?

[gautam]

You've hit the nail on the head, Shel!

The thing with the Long Tail of visibility is that you just need to be credible, and the visibility would not really be needed.

Over a long time due to conditioning of the mass market we've always believed that the people who are more popular are necessarily better. That does not hold true in an age where one can be relevant for merely a handful of people and make a difference in their lives.

[Dario Salvelli]

I'am agree with you, i don'like words as influence or popoluarity. :)

[shel israel]

Tom,

You have a memory like a non-partisan elephant. I used that line while debating my very good friend KD Paine at a SoftLetter conference that I suspect you produced.

[Erich Morisse]

In the case of popularity vs. influence, we often use popularity as a lazy proxy for authority or quality: if everyone else says it's ok, I'll give it a shot. So unfortunately, popularity is a factor into influence. Fortunately, it's by no means the only one.

Chris Brogan has a good post about the related, but not identical, topic: authority on the web.

[Chris Saad]

A great way to measure production side influence is Buzzlogic.

Measuring consumption side is a different story all together though.

[bejewell]

This is really a fascinating read. As a "mommy blogger," it's very easy to get wrapped up in the day-to-day numbers. I'm not sure how much influence a blog about silly family stuff or personal thoughts can really have, but I would so much rather have my words actually mean something to a few than reach many but touch none.

[Ken Stewart]

Shel, I'm new to your blog. Found my way here via FriendFeed and Louis Gray sharing your post.

I love your post. I have been working through this in my own right since I began blogging in February of this year.

While I am new, I quickly picked up on some tricks of SEO - just as you had said. It was fun to play with things at first.

As I changed topics, to suit my interest, I noticed very quickly I was gaining and loosing followers - leaving me in a sort of dead zone.

Then I became focused on building readership - blasting my message out. In the past 3 months, I figured out I had lost something... what was it?

It was the ability to influence and be influenced - by my community. I found my way back to actively participating in the community. I quit trying to be popular and just tried to meet people and have engaging conversations.

What I found is that the Web is not so hollow and empty as I once thought. There is much substance; from the young lady I know who lives just down the street to the friend I made in New Zealand.

The world can be smaller and fulfilling. I can take the things I learn online and share them with those I interact with. Likewise, I hope to share those things I learn in life with those new friends I have found my way to as well.

Again, thank you for your insight.

[Tawnya]

As someone relatively new to the world of blogging, I can say I've already seen this to be true. In the beginning, I was a slave to my statistics and working hard on SEO tricks to pull people in.

What I found, however, was that the people coming to my site weren't the readers I was hoping for - when I began to focus on things I cared about, rhyme or reason be damned, I started seeing meaningful numbers rise.

I'm still small fries now, but I'm doing it out of passion instead of a need for popularity.

I've poked around on your blog - great job. I enjoyed your article. Count me in as your latest subscriber; I imagine I'll be poking in from time to time, gleaning as much knowledge as possible. Thanks!

[Nathan Hangen]

I love it. The numbers do nothing but inflate egos, but influence is often unseen and unnoticed.

Sometimes you have to go against your own instincts to become popular, and I just don't agree with that methodology.

[Melanie Baker]

These measurement questions are precisely why we've labeled our analysis engagement, and why we work on spreading the gospel, so to speak, of what that means every day.

For a long time being "somebody" online meant high traffic and a good Technorati ranking. Even if the sources of that information were sometimes wobbly in terms of rigor and accuracy. But hey, when that's all you've got, that's what you count and call it measurement.

Fortunately, times change, and tools improve and evolve. If social media is going to be a significant part of the landscape now and in the future, then we need to account for it. If we can track more than just unique visitors to a site, then we need to do so. That said, it's still early days and I think in a few years we'll consider new ways of measuring now to be terribly primitive. But it's a good start.

What's interesting is that for an immediate hit or an ego boost, a short-term snapshot of engagement does show popularity. And if you like, you can choose to extend the engagement analysis to compare your current performance to your own past performance (apples to apples) or to others in the same niche or competitors or whomever you choose (apples to oranges).

Long-term, though, tracking engagement via clicks, pageviews, comments, trackbacks, bookmarks, tweets, etc. shows not only current popularity, but tracks influence over time. And I think time is the key metric that differentiates influence from popularity.

This tracking also not only shows whether influence grows, shrinks, or stays the same, but allows people to tie in what actions of theirs, e.g. blogging topics chosen, writing styles tried, posting frequency, etc. has led to changes in popularity and influence.

[Miles Sims]

Melanie - well said!

I agree that time plays a larger role in true engagement and should be more heavily weighed.

We've also seen that specific posts or topics will have a larger impact on getting people to sign-up even if they don't post a comment on that item. Knowing the types of content that drive growth can be a great way to maintain influence as well.

[Snazi]

Are spell checkers really that hard to use? How about showing a little respect for your audience?

[shel israel]

If you followed me on Twitter, you would know that the spell checker on Typepad is broken. But thanks for demonstrating your focus. Perhaps you would like to volunteer your time to proof read me and the other people who voluntarily spend time creating content for you to review.

[Susan Getgood]

Exactly.

[frank]

hey Shel ...

1st ... bummer that someone had to be a jerk about spelling. (not my strong suit) ...

but more on the topic ...

what i have found in my short time in the world of SM is that people who will take the time to chat with me or engage with me (even in short bursts here and there on Twitter or through responding to comments)are the ones that earn my respect and gain a level of influence over me. those that never talk with me earn very little respect and their influence is very limited.

obviously the 'big names' get my attention because of the name they have and what they have done to make their personal brand 'big' ... but if those people never talk with me my interest goes down over time. which is a fairly natural outcome.

The principal of treating others as you want to be treated definitely applies.

--
http://twitter.com/franswaa

[shel israel]

Frank, I find myself responding to you, because like this time, I find what you say is often interesting. Ultimately what I want from my social media activity is to have interesting conversations. Monologues in social media make no sense. They are not social. Thanks for be interesting soften.

[shel israel]

Frank,

Um that was supposed to be "so often," "not soften." Wait until the spelling Nazi sees that one.

[Robert Worstell]

This is the exact right thing to say - the perfect approach to social media: don't care about social media. Be yourself and your true friends will flock to your side.

Wish the corporations would find out what their "true voice" is - and then we'd have to endure a lot less advertisements from them. You see, their problem is that they think we think a certain way about certain things important to them. So they broadcast ads on a percentage basis. Trouble is - the rest of us have to endure those ads for that small percentage of people who would be loyal followers anyway.

TV was made lousy just by this point. And Internet videos - which are content-based and applicable - are more viable promotion. People searching for fruit don't want to hear about Apple's computers. And so the cost of ads go down as well as their frequency. Only searchers for Pepsi need to hear about Pepsi. (BTW, the cost of PPC goes up - but the big boys are paying/playing with a larger budget...)

Moral: build a great niche site and speak exactly what you've always wanted to say on it. - And then sell PPC ads for extra income!

[Laurent Nicolas (Alenty)]

Alenty has developed a system that computes the level of influence of a member within a community.
This level of influence does not only depends on popularity. It takes into account:
- time spent by readers reading what the influencer wrote
- level of influence of the readers
- comments writen by readers
- level of influence of authors of the comments
- time spent by readers of the comments
- level of influence of the readers of the comments
- grades of the articles writen by the influencer
- level of the people who gave the grades
- comments writen about the comments
- grades of the comments
So, really, influence is an infinite loop that takes into account all the possible interactions.
In Alenty's system, this level of influence is represented by a temperature (which is more fun than a kind of "value").

[Alex, aka SocialButterfly]

Amazing article and great point. Reminds me of what some say about high school....with the class president being a popularity contest. My question then back to you is: Who is better for the job? Some one who make be more popular or someone who holds more influence? I don't think either can exist in a bubble. It'd make an interesting thesis if you could some how quantify each term and then copare that to results or changes.

[cheap seo service]

i think big numbers are great. The bigger the better.