Why are you reading this? On writing for an audience
As lead blogger at AnnArbor.com, I get questions from our contributors about writing that are really different from any of the questions about writing that I had when I was learning how to write in school back in the days before the Internet. In addition to the usual questions about how to work the mechanics of our publishing system and how to format the text so that it looks decent, there's a whole layer of questions around the topic of how someone can be certain that anyone else will read what they just wrote.
Why are you reading this, and how can I make it more likely that you will read the next thing that I write? Here's a look at some of the ways that people find things online, and how you can place what you are writing in front of more of the people who are more likely to take the time to listen to what you have to say.
Write early, write often
The very first reason that you are reading this is because I wrote it. If you are writing for an audience, you have to hit the publish button before anyone can read it.
This may sound obvious, but it's more visible as a cause and effect when someone who has otherwise been a regular correspondent takes a week off. If you're not writing something new on a regular basis, you lose the opportunity to gain an audience for each new piece that you write. A steady pace, where you can predictably produce something on a weekly or even daily basis, is more likely to generate sustained interest than unpredictable and infrequent contributions.
When I sit down with contributors to look at traffic numbers, the first thing I look at is a simple time series of page views per day. The typical result, especially for someone just starting out, is that every day that they publish something there's a spike of interest and every day they don't publish something the traffic is much lower. There's some extended period of getting started encouragement necessary to get people going.
Tell your friends and colleagues
You have written something, and it has been published. Now what? It's time to tell your friends who are interested in what you write that it's time to take a look.
When we see articles that get a lot of readership, it's often the case that either the author or the readers of the article have passed it along to their friends. Some portion of that we can measure, e.g. by looking at how many readers have started from a page on Facebook or Twitter or one of many web-based e-mail clients before they look at a piece that has been published. In other cases, we notice incoming traffic from community news sites like Reddit, local and regional weblogs, and occasionally even another news organization. Mailing lists, especially the type of widely read and narrowly focused topical list that is kept on topic by a careful organizer, can be excellent ways to inform.
A writer builds up an audience not only of readers, but also of others who will pass along their work. Your friend who is a good source of interesting news is also a good person to keep informed of what you have just put together that he or she may want to share.
Be obvious for the search engines
Google, Bing, and other search engines use a cleverly tuned and constantly changing set of rules to evaluate whether a page on the Internet is relevant to the search query that someone just typed in. You don't have to understand those algorithms to know their effect. Writing to the audience of everyone who is searching right now for an answer to their question means that what you just put on the page will be visible to the whole world.
It's always worth writing with one eye on how you would search for what you just wrote. If you have a well-written piece with a carefully chosen headline and a crisp beginning, you can extend the readership of the article far beyond any of your immediate acquaintances to the broader Internet that is searching for just exactly what you just wrote down.
My favorite recent example of this is a short article I wrote late at night as Comcast was having network troubles. Entitled Widespread Comcast outage, and how to route around it, it was four short paragraphs about the problems that Comcast was having at the time and how you might go about working around those problems so that you could get back on the network.
Much to my surprise, this post got a huge amount of traffic, some of which was from people who had put the single word "comcast" into their search box. The writing was timely, concise, accurate and answered some current problem, and it had also been shared widely on Facebook and Twitter. The net effect was to bring in thousands of new readers.
Not everything you write will be front page material on every search engine. What you do want to be able to do, however, is make sure that you know what to type into a search engine to find everything you want to share. At the very minimum, the combination of your name and the topic you wrote about should return all relevant and interesting results. When I write about sledding, I expect that the search for Vielmetti sledding will come back with interesting results on the whole page.
Be kind to your editor
I field frequent questions about how someone should write something so that it will be featured on the AnnArbor.com site. The answer, I fear, is both more and less complicated than it needs to be.
Every piece that's published is put into a category at the time you write it. Some of those categories (news, sports, crime) get a lot of traffic on their own, but not all of them do. Classification is a way to signal to the world and to the editorial staff what something is about, and it is a way to make certain that the person who is in charge of that category gets a chance to look at it and to see whether it's worthwhile to promote further.
There's an editorial process that goes on constantly which looks for some parts of the stream of news to feature on the home page. Without trying to reconstruct an entire journalism profession in a single sentence, this process balances newsworthiness, timeliness, popularity, interestingness, and other ineffable qualities that as a blogger I am unlikely to ever really understand.
The answer then is simple, even though I can't explain the decision process. If what you write is likely to be interesting to the editor's sense of who the AnnArbor.com readership is, it's more likely to be featured. Your demonstrated and repeated ability to bring an audience to what you write increases the chance that an editor will look at the next piece favorably.
Causal inference in social networks, and why it is hard
I was at a very interesting talk by Carnegie Mellon University professor Cosma Shalizi, who was back visiting the University of Michigan Complex Systems group this week. Entitled Homophily, Contagion, Confounding: Pick Any Three, it provides a suitably complex and confounding answer to the question of why you are reading this post.
Cosma notes that there are two sets of reasons that might provoke one person to do what another person does. There might be contagion, in the sense that one person convinces his friend to do something else by force of persuasion, akin to a contagious disease passed on through friendship. The second is homophily, usually expressed as "birds of a feather flock together", where knowing that both people are interested in the same topic makes it more likely that they will both react in the same way to the same stimulus.
He gives the example of Irene and Joey, where we want to know, if Joey jumps off a bridge, whether Irene will follow. Joey may have persuased Irene by force of example (social contagion), or he may have infected Irene with the same brain-curdling parasite that caused him to jump (biological contagion). Or Irene and Joey may have met in a bridge-jumping club (manifest homophily) or a roller-coaster thrill daredevil club (latent homophily) where both of them already had bridge-jumping tendencies.
Cosma's confounding conclusion, based on models of how you could observe Irene and Joey's behavior, is that it's impossible to tell just by external observation which of those two sets of reasons explain behavior. (Read the paper for all of the details.) The over-simplifying further conclusion is that you also couldn't really know what caused you to read this article. It may be that you're interested in the same things I'm interested in, and it may be because we know each other, but trying to tease apart which is which is not easy.
A conclusion of sorts, and some links
Be interested in something, and have friends who know about and share your interest, and be persistent; if you can manage each of these, it may be impossible to know which of those makes the difference in whether any individual thing you do will be noticed and recognized.
Causal Analysis in Theory and Practice, a weblog by UCLA philosopher Judea Pearl.
Is height contagious? looks at implausible effects of social networks, based on a paper by Ethan Cohen-Cole and Jason M Fletcher that "proves" that height, acne, and headaches are all contagious.
The University of Michigan's Center for the Study of Complex Systems (CSCS) is a broadly interdisciplinary program in the College of Literature, Science and the Arts (LSA) at the University of Michigan in Ann Arbor, Michigan. Our mission is to encourage and facilitate research and education in the general area of nonlinear, dynamical and adaptive systems. Its occasional seminar series is fascinating, confounding, and determinedly relentless in disabusing you of the notion that there is any simple explanation for anything.
Edward Vielmetti goes to interesting seminars rather less often than he should for AnnArbor.com.