A Month Of Traffic To AVC
It's 5am and I've got to catch an early flight to Toronto and will spend the day on the road, returning around 8pm. So I don't have much time to blog today.
I thought I'd share what a month of traffic to AVC looks like. Monthly visits to AVC average 300k and page views average 400k, but how you get there is always fascinating to me. It's not 10k visits per day every day. Far from it.
Some days are slow with 5k visits or less. Some days are big, like yesterday, with almost 20k visits. In fact there are very few days where the daily visits are near the average. Blogging traffic depends on lot on the post that day and how it reverberates around the social web.
Comments (Archived):
any correlation to popular posts?
Yes. Very much so
I’m curious about the relationship between popularity (traffic) and # of comments and even # of commenters Fred.This may be more magic than science but I’d be interested in your thinking on this. There are may be some correlations that define an engagement graph of sorts for avc.com.And BTW…these are remarkable numbers. Congrats.
They are correlated for sure but not tightly. Sometimes a comment sparks adebate and that can drive as many as twenty to thirty comments. That happenssomewhat randomly
Figured as much.There’s knowledge in that engagement data that we don’t know how to plumb or use yet though.
I’m guessing Martin Hill’s comment in your “Android (continued)” post was one of those exceptions (I might have sent a tiny bit of additional readers over there with this Seeking Alpha article that referenced it, “The Battle for Dominance among Mobile Operating Systems”).
might be interesting to see if time spent correlates more tightly, although other things like length of post and whether you link to a video probably a factor too
that’s an excellent point, and it shows how blogs are different from standard websites.there’s a lot of action here that happens solely within Disqus, which has a lot of slick features that make each post perform more like a web app. All that activity may or may not be accurately reflected within web analytics tools that are measuring “page views”.
“…each post perform more like a web app”Interesting. Nailing the ‘why’ of the gestalt of Disqus is an ongoing quest. This is a good stab.Re: whether these are reflected in analytics, I don’t think so. There are two sides to this, On one side, the value of the comments points beyond pure traffic to true interest connections.But it would be interesting to throw a heat map on Fred’s blog and see how much of the traffic spends time on the comments and the onsite or oncomments time. My assumption is that the Disqus thread is the stage and magnet for drawing traffic. Just a hunch though.
I co-run a development and deep analytics firm, and can say that on the technical side, most analytics still just rely on old-fashioned server calls, and most of the in-page action (which Disqus slickly does) gets under-represented. There are ways to engineer around this, but the decision points aren’t so straight-forward.That uncertainty is due to the side you mention: that it’s tough to really measure attention and engagement, which is a leap beyond pageview numbers.
Because. I can’t answer below: what are some more common issues that you see whenfor creating tools to get the metrics when you have a difficulties measuring (eg, as you mentioned with disqus)? I’m thinking that certain questions one could have/get answers to could create more technical issues than others, which is why you’d have to do more engineering.
well, there’s reading a webpage/blog post. And there’s leaving a comment. And then there’s multi-way discussions like we are having right now, deciding to take time away from our busy day and put thought into.I can’t think of a deeper form of (non-monetary) engagement than that. So a well thought-out metric would try to take that into account, rather than just page views which are a relatively stale number.
hello from kathmandu!
When did you guys add an “h” to the name of your city? I must have missed that. But Wikipedia redirects from “Katmandu” so I guess it happened.
Can I ask what you use to get these numbers? What’s the difference you see between server logs, comscore, alexa etc?
This is sitemeter which I’ve used since day one. I also use Google analyticsand chartbeat. I’m a numbers junkie
the only bad part of my web publishing is that I get addicted to stats. and readers. did the governor read my blog today? if not, what did I do wrong?i use analytics and sitemeter. what does chartbeat bring to the table?
Congratulations. You can get lost looking for patterns I bet, though Twitter definately figures in big time.
What’s in Toronto, kik?I know there’s no football team up there (Maricel Dareus, Bills!). Can’t wait to see if we snag a QB today.
Yes. Kik
What is the yellow and red shade represent? (Anything to do with direct and through other sites? because the red shade looks more consistent (direct visitors) than the yellow one’s…just a wild guess!!).Continent wise data would be interesting to see how popular you are outside U.S. Any big numbers from Asia (china, japan, Australia)?
Speaks volume about the fact that it’s the type of Content you write that’s a primary factor for driving the traffic. There’s probably a baseline that’s held because of who you are, but then, above that, it’s the content.
It’s a great community that I look forward to visiting everyday!!Spikes have to be that crazed “marketing’ weekend.
That was an amazing post yesterday – generated lots of noteworthy comments, too!
Enjoy Toronto. It’s cool and windy today.
running a regression to predict your traffic would be kind of fun, actually.assume that it lags across time and include a dummy-variable for weekend should get you a decent Rsquared.
what would be the independent variable there?
twitter mentions?
Maybe the “mode” would be an interesting stat to look at (if you had a classification like 10-11k per day). People tend to ignore that in statistical analysis.
I am approaching 2K daily myself. Hit a high of 3K months back.
I’d say you’re an outlier when comparing stats to most blogs. In your circle of bloggers, they might see similar things. However, most bloggers don’t have days that are influenced by that specific post. They are generally close to their average and views are usually very cyclical (ie. weekends are lower traffic). Point being that Fred Wilson creates a lot more “reverb” than most bloggers, because Fred Wilson’s readers are great at reverberating. Most bloggers instead follow a more slow growth path using a combination of online marketing tools with the occasional spike because of some interesting event or post. This said from someone who runs a couple blog networks.I appreciate you sharing the stats. They’re definitely interesting. If you’re a stats junkie, you should try out the WP.com stats plugin. It’s my favorite for tracking the stats on my blogs.
Very interesting to see these stats!The division of visitors during the week also depends greatly on what you blog about.I run a niche blog (about energy certification for houses) for a local market with +- 2k visitors per month. Most of my visitors come on monday, probably because they prefer googling than working monday morning. The least of my visitors come on weekends while they are out shopping and relaxing.
Controversy sells.Re: commenting, so do topics people can personally relate to and not be technically ‘wrong’ in commenting.
OT, but AVCers who haven’t seen the news might be interested to learn that Tereza’s company is now one of the finalists in Business Insider’s Startup 2011 contest.Kudos to Tereza for that (and for writing a pithy enough bio that it fits in the space Disqus provides).
I <3 @davepinsen! Thanks, dude!Hey — anyone AVCers who haven”t been on honestlynow.com yet, go do it NOW……or I’ll personally find you and make up highly personal and embarrassing questions about you that should be asking about yourself. 😉 j/kAnd YES we know there’s tons of stuff that needs improving, and we’re working on ’em. But in the meanwhile, there’s plenty of fun to have by GETTING HONEST!xoxoT.
Congrats and felicidades, Tereza!
I wasn’t aware, thanks for the tip & good work to Tereza and team.
I didn’t think she’d toot her own horn here, so I figured I’d do it for her. Plus, she was kind enough to re-tweet some positive news of mine yesterday.By the way Mark, you (and Arnold Waldstein) also deserve kudos for writing descriptive Disqus bios that fit the space allotted for them.
I think you are right about the more general the topic ostensibly (like marketing) the broader the population that is comfortable in responding.Shock blogging works but gets old quickly.
very good.
Not surprised by this at all — I have an MBA so I almost never even bother to open your MBA Monday posts. Although they are well written, I don’t learn anything new from them. However, I guess I am not representative of any particular class of reader, otherwise you’d expect to see that reflected in the traffic for Mondays.
what should i write about tomorrow that you might now know about yet?
Very good
I don’t know if you look at your feedburner stats much, but there are likely a lot of people that read the blog from their RSS reader like Google Reader and then only come over when there are a lot of comments.20,000 visits to 300+ comments is an interesting ratio to me. I wonder what the norm is.
i do check feedburner every month or soi get about 7500 RSS “visits” a day