Chuqui 3.0

Can Pay-For-Performance Improve The Quality Of Content On The Web?

A couple of recent postings on Gawker moving to a "pay for performance" model:

Can Pay-For-Performance Improve The Quality Of Content On The Web? - Publishing 2.0:

Nick Denton and Gawker Media are wrestling with the problem of content quality on the web — specifically, how to give bloggers incentives to create content that drives traffic based on quality rather than quantity. Gawker has announced that incentive pay for its bloggers will now be based entirely on the number of page views that each blogger’s posts generate, rather than on the total number of posts a blogger writes.

Pay for traffic: Incentive or distortion? - - mathewingram.com/work:

The key question, of course, is whether rewarding bloggers for traffic is a good thing or a bad thing. One argument is that “incentivizing” bloggers to boost their traffic encourages them to make their posts more sensational, and will lead to them writing about nothing but Britney Spears or whatever they think people will be looking for, instead of deep and thought-provoking posts about serious issues. This is similar to the argument about people writing just because they want to show up on Techmeme.

The opposite argument is that it’s good to give writers a stake in the success of their blogs, something that encourages them to take an interest in their community. Will that encourage them to “sell out?” Perhaps. But maybe it will also encourage them to respond to comments, link to others who are discussing the same issues, and so on. Even former Gawker editor Choire Sicha thinks it’s not such a bad idea.

I have to admit that I don't like this model. I considered it for something I'm working on and ended up rejecting it in favor of a different model. It works -- and can work well -- but I think it has side effects that I just didn't like for what I am hoping to do.

It's a model that works well for sites like Engadget -- where there's not a lot of deep content or significant writing, but instead acts more as a link blog; you don't read Engadget for them to inform you, but for them to post links to other sites with interesting things on it. In a case like that -- the better you are at generating messages that include links that draw users, great.

But for sites hoping to get writers to invest in more thoughtful writing (instead of what I call the "engadget house style", that of being two paragraphs, a semi-related but catchy photo, a hot headline and a smirk), I think pay for performance rewards the wrong behaviors. It encourages LOTS of short and quick postings, not more thoughtful, longer pieces. It encourages thinking about maximizing page views instead of developing audience, and it encourages writers to promote themselves rather than the site in general.

My current plan -- heavily subject to change -- instead tries to look at the authors as members of a community, that community being those that contribute to a site. I want to build a financial model that encourages those members to support and help each other and benefit from the success of the community (and site), not just encourage them to fight for visibility and pageviews.

This model presumes more of a traditional magazine format -- fewer, longer articles, more use of freelance writing than "staff" writers, than what Gawker does (hey, someone has to create content for people like Engadget to point to!), but in that situation, pageviews is not king, nor is driving lots of daily traffic. Instead, I think the key is building a strong regular audience and a long-term draw of visitors to the site. Where Endgadget seems to need to depend on however many views they can get within 72 hours of posting a page, I'm looking at creating a place where pages continue to draw for weeks, or months, or years -- a slow, stead growth in traffic and interest over time.

So what I'm doing is creating a three tier payment system. Content creators live in one of those three tiers, and get paid depending on the revenues of the site and what tier they're in.

First, nobody gets paid until the site pays expenses... that, I guess, should be a given...

Tier one: founders/editors -- the people who do the day to day work, keep the site running, handle acquisition and all of the business aspects of the site.

Tier two: contributing editors -- people who are considering regular contributors to the site; generally this are people who've proven they can contribute on average 1 acceptable piece per month over a period of time.

Tier three: contributors -- people who contribute occasional articles, or a single piece.

Revenues are split 40/30/30: 40% of revenues (after expenses) gets split by the founder/editor group because they're not only writing, they're doing all of the other day to day crap that keeps the site running and useful; they need compensation for all of that (and the more they write, the less it's worth in some ways, because their share doesn't directly tie back to the writing, it's "part of the job". but founders are expected and expecting that sort of thing.

30% goes to the contributing editors, and they get paid based on the three months surrounding their article publication, so if it's published in the March accounting period, they get paid out of the 30% pot for February, march and April. Since part of the reason they do this is because they're regular writers, they basically get a monthly payment for contributing to the site -- and it comes from what ought to be a smaller number of writesr, so the share of that 30% is hopefully larger.

30% goes to the writers, in the month their piece is published.

Those percentages may change. there are all sorts of details to work out, such as "what if something's published on the 29th of a month?" (answer: I dunno yet). It may be the reporting period is whatever accounting month has most of those 30 days in it.

One issue I'm still grappling with: an article published that draws good revenue for a year or more might not benefit the author at all. On one hand, I'd like to hope this encourages the author to continue contributing, on the other hand, I think a successful article deserves some kind of royalty payment. I haven't decided how I plan on resolving this yet. Maybe set aside 10% of revenues to royalty authors of articles that "earn out" in some way with continuing traffic.

My main goal here is to create something that's equitable to all involved -- but encourages people to think in terms of making the site successful, not just promoting their own articles or themselves to the detriment of the site. Also a model that looks at long-term readership as a revenue stream, not hot and heavy pageviews that I think encourage "bitchslap journalism", or "people magazine on the web" type stuff. It's endemic on the founder/editor group to not tolerate "freeloaders", people getting revenue but not contributing any more, and to make high quality choices in content approval/acquisition.

But I think it's a much better model than pay for performance, especially if you're trying to encourage deeper, more thoughtful writing. Think of it as the "Esquire" model. At least, I'm hoping so...

(and yes, it's all been on hold for a while, thanks to the new job-that-pays-the-rent. But it's being dusted off again, now that I've more or less finished the ramp-up stage and gotten the worst crises under control...)

Digg Traffic Has Questionable Value For Most Niche Publishers - Publishing 2.0

Digg Traffic Has Questionable Value For Most Niche Publishers - Publishing 2.0:

What I’ve just said might come as a surprise to many niche blogs and traditional media publishers who play the volume game when it comes to selling the value of their online audiences. Many tech blogs report audiences measured in shear volume, i.e. hundreds of thousands or north of 1 million, and those numbers are goosed by avalanche traffic from Digg.

The problem is that from a trade publishing ad sales perspective, there’s limited value in raw volume. Trade publishing is all about targeting and focus. Most Digg users are still in high school or college, and are therefore unlikely to have any B2B purchasing power, so they have questionable value to the B2B advertisers on many tech blogs — particular those that cover the BUSINESS of technology.

Of course, tech blogs have CORE audiences that are highly relevant to their advertisers, but the traffic from Digg adds little if anything to that value, and in fact can potentially dilute the value for publishers that sell on a CPM basis.

This is not to say that there aren’t blogs and other online publications that benefit greatly from Digg traffic — for gadget and gaming blogs, Digg traffic is hugely valuable, because Digg users are certainly interested in gadgets and games, and they probably buy them in large quantities. Some types of political stories do well on Digg, as do stories of the weird.

Scott Carp talks about the value of Digg traffic -- and finds very little value to it.

This is something I talked about a while back, and came to the same conclusion:

Chuqui 3.0: does digg or reddit matter?:

Well -- I'd been experimenting with a "digg this" feature on typepad. It's gone. I see no reason to encourage those folks to come visit my blog. They generated a LOT of page views, but most of the constructive things I saw from the viewing tsunami I can attribute to reddit. Digg definitely bumped my page views, but didn't generate traffic elsewhere on the site and (to the best of my interpretation) limited subscription increases -- and definitely not increases to warrant putting up with all that other stuff.

I'd disagree with one thing Carp says -- I doubt that Digg traffic is very valuable even for the niche blogs that match up with Digg demographics, because Digg traffic brings along very little EXCEPT raw page view numbers, which are a resource suck and really have no value to any site except those that still include page counters on things (haven't we grown beyond that, folks?); Digg users don't explore a site, they don't subscribe to a site, they don't click on advertising on a site. They see a link on Digg, they click through, they glance, sometimes they drop off a snide comment, rarely they'll link to you on their site, and then they're gone.

If you're blogging for anything other than enjoying hearing yourself type, you're blogging for either revenue or for a readership. Digg generates neither; the users that come across don't subscribe to your feed, and they don't generate advertising revenue. A store full of window shopppers may look busy, but the store's not making money, and that crowd may well deter real shoppers from coming inside. That's the model Digg brings to a blog. As far as I can tell, there's no benefit to a blog by being linked to from Digg; it's all about Digg benefitting from your blog. It doesn't return any favors here...

I stopped encouraging Digg back in August. I haven't missed it. And my feed numbers are up significantly without them...

Evaporative cooling of group beliefs

Evaporative cooling of group beliefs:

Over on Overcoming Bias there was a great post called "Evaporative cooling of group beliefs" where the author talks about how ejecting outliers moves the group's average position towards the other extreme.

[...]

My own theory of Internet moderation is that you have to be willing to exclude trolls and spam to get a conversation going. You must even be willing to exclude kindly but technically uninformed folks from technical mailing lists if you want to get any work done. A genuinely open conversation on the Internet degenerates fast.

[...]

It's interesting to compare this to the techniques Theresa Nielsen Hayden uses on her "Making Light" blog and on Boing Boing comments. There's an art to building online communities that nobody has yet well documented.

Does it move it towards the other extreme, or merely back away from the far edges of the bell curve? A move towards the other extreme can also be a move towards the middle, depending on where the starting point is.

To me, the reality of community management is that early on in the process, a moderator's policies, style and attitude shapes the community, because it'll attract certain people and attitudes and discourage (or ban) others. Once a community is established, it really needs to dictate policy and the administrator's job is to steer the group in the direction it wants to be steered, and where necessary protect it from those who put personal interests above the needs and wants of the group (trolls are, ultimately, people who are more interested in garnering attention than contributing to the group, or in some cases, simply people who insist on "winning", which excludes others from the option of "winning", however the community defines that)

I admit to not being a big fan of how Teresa manages her communities, but I'll be the first to say that it works for her and her groups. On the other hand, I find those groups very "echo chamber"-ish and narrow in viewpoint and not terribly tolerant of views that don't fit the common worldview. Well, most communities are like that, but those are even more so. But they're happy with it, and that's what matters. What I think doesn't, since I'm not involved in them.

At one point I thought it might be fun to write "the book" on community management, whatever that is. Later on, I thought maybe building a wiki or some kind of community for community managers to write an online guide might work -- then I realize what it would take to be community manager for a community of community managers, and I went and put a washcloth over my eyes on the couch until the thought passed...

I still don't think I've seen "the book" on all of this, or even if we're at the point where it can be rationally written. There are so many right answers -- each depending on a situation and the personalities involved; and for every situation where a community policy works -- it'll fail miserably in a dozen other situations with different people and different needs. So maybe it's best not to put all this in stone. Or paper. Or whatever...

Hey, where’s your journalism licence? - - mathewingram.com/work

Hey, where’s your journalism licence? - - mathewingram.com/work:

As far as Hazinski is concerned, only “real” journalists can make sure that the citizen kind don’t go around making things up and not playing by the rules. As he puts it:

“While it has its place, the reality is it really isn’t journalism at all, and it opens up information flow to the strong probability of fraud and abuse. The news industry should find some way to monitor and regulate this new trend.”

Did you get that last part? The news industry should find some way to “monitor and regulate” this new trend. With what, Dave? A central tribunal of some kind that can pass judgment on who has committed acts of journalism and who hasn’t?

Heck, let's just extend the current monitoring and regulation that the industry does on itself to bloggers. That should solve the problem, right?

After tall, we have no worries about fraud and abuse in the "real" journalism world. Well, as long as we don't think about people like Jayson Blair, publications that bias and slant the news to meet specific agendas (hell, that goes back to the first newspaper, but we can stop at William Randoph Hearst -- but it carries through to today). We won't even mention the Weekly World News (the world's only reliable newspaper! Hey, if it's in print, it has to be true!)

Come to think of it, we already DO have bloggers covered by the same monitoring and regulation as the journalism industry... Maybe if journalists set up a central organization of self policing first, then bloggers can join in later after it proves itself to work?

Oh, wait. that's DIIIFERENT....

SezWho Comes to Venture Chronicles : Venture Chronicles

SezWho Comes to Venture Chronicles : Venture Chronicles:

I installed SezWho on this blog a few days ago and am really liking it.

One of the most significant PITA challenges in the blogosphere is managing comments. You can use services like CoComment, but my experience with their Firefox extension leaves a lot to be desired (it froze my browser frequently) so I’m not inclined to recommend it. Furthermore, CoComment does not rate/rank comments or plug into my blog to track other people’s comments.

SezWho does two things really well. First and foremost, their Wordpress plugin drops a "was this comment helpful to you?" prompt on every comment. This binary yes/no is applied in two ways, first the aggregate ratings are presented in dashboard in my SezWho account, which gives me a unified view of what comments are posting and what people are ranking their usefulness as.

Secondly, as more people rate a comment up/down the ranking increases or decreases accordingly, which helps visitors pick the comments that are truly useful. On this blog I am not fortunate enough to have hundreds of comments on any given post, but on high traffic blogs I can see how this would help me as a visitor increase the signal-to-noise ratio.

Sezwho looks to be doing well so far. I'm hoping to see Typepad support it; otherwise, I'll likely end up moving to Wordpress (or maybe Drupal), and I'm not looking forward to moving everything again just yet...

Total versus Active (or Dormant) Blogs

Jaffe Juice: Total versus Active (or Dormant) Blogs:

According to Technorati, there are over 108.9 million blogs that are now being tracked. I wonder how many are active (and also, what the definition of active is)?

I've asked for that in the past myself. (like here, in 2005).

Or as I noted back in July:

Chuqui 3.0: My Blogging Reality Check » Webomatica - tech, movies, music blog:

According to the Technorati page today....

Currently tracking 93.8 million blogs and over 250 million pieces of tagged social media.

So, there are 93.8 million blogs. my blog has a technorati authority of 102, and a rank of 51,458.

Okay, what's wrong with this picture?

my blog, which has 500ish feed subscribers and a few hundred page views a day, is better than 93 million blogs out there? it's on of the top 50,000 blogs out of 100 MILLION?

it's worse than that. A couple of months ago, because I was testing some things, I created a typepad blog, made two postings to it (of the "hello world" variety), and then never touched it again. Within a few days, technorati ranked it about the millionth rank. For a block with two postings, no subscribers, no links in or out...

As of today, my feed readership is between 550 and 600. My Technorati auhority is 117, my rank is 54,800ish. All things considered, my blogs are small, very much personal.

Yet, do the math. 100 MILLION blogs, and mine is the 50,000th largest in the world?

Or do your own test. create an empty blog, register it with technorati, post a couple of test messages, and do nothing else. Don't advertise it, don't point to it, don't create content. you'll likely find that Technorati will throw it somewhere around 1mm in authority.

Which indicates to me that if a blog with no links, no content and no promotion is beating out all of those other 100 million blogs, then those 100 million blogs are pretty much like that empty one. Someone please prove me wrong. They're eitehr abandoned, stillborn, or some kind of spam blog. Actually the spam blogs probably rank HIGHER than the empty blog...

My guess is that there are probably 100-200,000 blogs active and with an active readership LARGER than the set that contains parents, siblings, spouses, best friends, and the cheerleader clique and the high school. Maybe 500-700,000 active blogs at all.

I think this is a place where Technorati has played fast and loose with the numbers becaues it made great PR, and has never come clean about what the numbers REALLY mean. I see the advertising spam blogs that scrape my content appear in technorati -- which is interesting, because I also have various scrapes in google blog search and blogpulse, and they don't show up there, generally. Hmm.

Somewhere along the line, Technorati made a decision that it was good -- for Technorati -- to be able to puff these really huge numbers. It's created a false impression on blogging, and makes it hard for me to believe anything they say these days; makes it hard to trust their numbers -- it's part of the reason I find myself using tehm less and less, and I've more or less decided to remove references to them from the blog when I get around to it.

But if you're talking about PUBLIC blogs: non-password protected, open for anyone to read, with real content, and more readers than Mom and their pretend girlfriend? Well under a millino. Depending on how you want to define "dormant" and "readers", we can go from probably 200K to 800K. Those are the numbers I tend to use these days.

And heck, being 50,000th out of 800,000; that's still not so bad, given I haven't particularly tried or cared to be a so-called "A lister". but when I looked at it and realized that I was 50,000th out of ONE HUNDRED FREAKING MILLION BLOGS, I realized something simple -- I realized those numbers were both worthless and wrong.

Update: Kevin Marks @ Epeus' Epigone disagrees with me.

So what of Chuq's contention that the other 96 million blogs with no links are "abandoned, stillborn, or some kind of spam blog" ? A lot of blogs are made for specific events, and don't need further posts adding (and so may not have been linked to in the last 180 days), and a lot have a low posting rate, sure. But, as Dave Winer once said, by that measure every book and magazine article is abandoned. Not all blogs are interested in links - many are personal journals or for a small group of friends to read, and achieving those goals well.

So Chuq's accusations of "playing fast and loose wiht the numbers" are really his misunderstandings of Long Tail distributions. In the Blogosphere, like Lake Woebegone, everyone really is above average.

He may be right. I don't think there's data to prove it one way or another. But -- to be honest -- I find the idea that it's okay to say "everyone is above average" just wrong, on many levels. If the data you have tells you that everyone is above average, that's really telling you your data is crap and shouldn't be trusted.

death to blog categories!

So....

One of the things I've been looking at recently is how the information on my blogs is laid out, and how that impacts things like SEO and how people can find things on the blog.

One of the things I simply don't understand, not even remotely, is why standard blogging systems store each blog entry three times.

Take, for instance, this article, currently visible on the home page of chuqui 3.0. Igonring the copy on the home page, though, the entire article is also published two other places.

One is on the date archive page. The other is on the category page.

Fortunately, the permalinks all point to the individual entries, but the more I think about this, the more I have to ask myself a question:

Forever god, WHY?

In researching what people are finding on my site off the search engine, I've come to realize this standard format creates a couple of problems.

First, you're competing with yourself. The same content is on multiple pages on your site, all of which are going to get spidered into search engines like Google. you are, effectively, diluting your content; while pagerank helps here and most links incoming will link into your individual page, this doesn't help the find-ability of a given article. Especially because...

Second, those collections cause false search hits, which bring users to your site that don't find what they want. That's not a way to encourage them to return. That's because on those date-based or category-based pages, all of the articles are spidered together. that means if a user of a search engine asks something like "Where is Elvis Hiding?", and you wrote an article one day on Elvis Stojko, and then four days later wrote a piece on how your cat insists on hiding your car keys in the litter box -- Google may well send them to your site, because you do, after all, have a page on Elvis Hiding (and where are those damn car keys?).

These kind of pages also seem to make adsense a bit crazy, so they can't have good advertising returns.

All of this seems like an anachronism, like the early days of blogging when we had those silly calendars on the main page -- remember those?

So I'm seriously thinking about how to change this. My thought: categories are just a physical manifestation of tags, so either do away with them completely and rewrite the templates so that categories actually are coded into the web pages as tags; that's let you use categories as your set of "standard tags" you code onto articles, and then add custom tags as you wish via marsedit or ecto or whatever. This seems pretty reasonable, but I need to build a prototype and see if I like it for real -- but I'm curious what others think and whether they find these massive pages of glops of semi-related text useful...

On the date-based archives, it's more problematic. For one thing, you need pages that link to those individual pages so that search engines can find them; at least one set of historical archive pages have to exist (and given that, I wonder what's more useful, date or tag/category?). but there's no reason why the entire article has to live there. Instead, treat them as a Table of Contents, and build an archive template with the article info and links and metadata, and a limited except (50 words? 100?) for someone browsing who needs some context. But not too much context, for fear of confusing or diluting the information.

I'm a firm believer that information should live in one specific place, not be duplicated out into lots of places without a good reason (think of it as database normalization for blogs....); it's my feeling that the reason blogs are structured this way is because that's the way it was done in the early days, and nobody's really sat back and rethought the implications here, especially with the growth of tagging and hte importance of search engines in the larger scheme of things.

So now I'm taking a look at this, and tossing it out to see what other people think. I think, at the least, we can do away with those category archives and build a more efficient date-based archive, and clean things up without any real negatives. That seems to match what I'm seeing in terms of blog usage, both by users coming directly to the site, users coming in via links on other sites, via the feed, or through searches.

Am I missing something here? Or does this sound useful to others?

Nike Develops Sneaker Exclusively For Native Americans, Helps Their Health

Nike Develops Sneaker Exclusively For Native Americans, Helps Their Health:

As it turns out, Nike’s not going to make any outrageous profit from the sneaker; it was designed to improve Nike’s image and standing amongst, you know, people.

Hey Nike, how ’bout a blogger shoe? One that makes running around press events painless and continuously pumps caffeine into your bloodstream. I’d buy it (or get John to buy it for us).

Hey, Nike, how about a shoe that kicks people in the nuts if they can only see things as opportunities for snotty remarks that they think make them look smart and cool?

folks, Spy magazine folded years ago... Does everything need to be an excuse for a snide remark and a "I'm smarter than you are" attitude?

Guess so..

blog postings need an optional expiration date...

I've been doing a lot of work trying to re-architect my online presence (aka, beat these damn blogs into some kind of shape so they better represent what I want them to be and who I am -- and so you can bloody find stuff by putting the stuff where it's expected....), and one thing I've come to realize is that blogging systems need to implement an expiration date for postings.

Optional, of course. And I realize that some folks think that every page taht was ever posted needs to be posted for ever, and that URLs can never change -- but get over it. Not ever happening, and frankly, shouldn't.

take this posting for instance. Please. 30 days from now, who cares? The web would frankly be a better place, or at least less cluttered, if we all got our giggle out, and then my link to Global Nerdy goes away...

(okay, too much fun. the ads by google showing up on this page all are about finding or buying laptops really, really cheap.... online fences? nah)

In all seriousness, some stuff ought to be kept around and links are a lifeblood of identifying and sharing useful info, but there's a large set of blog entries out there that are, in reality, transient. Will anyone really care two years from now that there were room available this christmas at yosemite? No, and in fact, that posting might just confuse someone looking for room reservations through Google or a search engine. So I should remember to go delete that sometime in the future.

When I started getting serious about Dare2Thrive, and realizing I wanted/needed to rearchitect my online presence (more on that sooner, or later... or whatever), I did some digging through my online "me" -- there are over 3,000 individual postings, plus 2,000+ flickr photos, plus god knows how much (and I don't want to know) in Google groups from my years on USENET or on various mailing list archives hanging out around there. tens of thousands of pages.

but even in the 5,000 or so pages I directly control going back to, oh, 2001 or so, my sampling shows that about 20-25% of them either point to something dead and broken (and therefore should be deleted or fixed), or are silly, or no longer relevant, or no longer represent an opinion I want to portray, or maybe a fact that's been proven wrong.

Yet it's all still out there. And blogs are great at helping you POST content, but lousy at helping you manage it. And we, as bloggers, are universally lazy about going back and fixing it once it's in there (and I don't blame us...).

One way to improve this would be to allow us to throw an expiration date on a posting. That'd be fairly easy for a site like Typepad: simply make it an option to say "kill this on this date, or N days from now"; when it goes pumpkin, Typepad deletes it for you.

Purely optional, but it'd give us that ability to say "this really is only of interest for a month...." -- and that'd leave us with cleaner, easier to use sites and fewer garbage links for the search engines to spider and try to make sense of...

Expiration dates aren't world-changing things -- but they'd make things a little better, no?

(another thing I've been thinking on is how to track down and handle dead links. Maybe it makes sense to build a little link-tracker system and use it instead of direct URLs -- could allow me some visibility into what people are clicking. I've looked at some systems to do dead link testing, and haven't found one I like yet (suggestions?) -- down this road lies serious analytics and reporting, though, so it seems to me this ought to be part of another existing package, like Google Analytics or Feedburner (which do the tracking parts, not the 404 testing part... sigh).

How do bloggers handle this, anyway? by ignoring the problem, right?

does digg or reddit matter?

A few weeks ago, when I did my Adobe rant, it ended up on reddit, and then later on picked up by digg. that created a huge traffic spike and some discussion across the net.

I was curious what kind of impact (positive or negative) this kind of spike might have over the long term. Now that it's abated and I've had some time to see what the new "steady state" is here, some observations for your enjoyment (or kvetching).

It was useful the way the traffic hit, because reddit had featured the piece about 24 hours before digg picked it up, so I have some ability to judge the sites against each other. Total traffic for the period was over 60,000 page hits.

Some base stats: feed subscribers before averaged about 440; today, about 475. That's up about 8%.

Feed-based reads are roughly double what they were; the people subscribed to the feed are reading it more actively than they were before.

non-feed traffic is up significantly -- where I was seeing ~600 page views a day, it's now 1000-1100. A lot of that is NOT to the page hit by digg and reddit, either; and I"m not sure why, unless this whole shebang simply raised my site's presence in Google so it's higher in the search results.

Of the spike, about 35% of the traffic came from reddit, 45% from digg, the rest from links from other sites, most of those links originated from readers from reddit or digg.

Only about 1% of the readers looked at ANY page other than the one linked to. It's clear that "digg is my home page", and their interest isn't exploring, it's pop in, pop out. digg users were a lot worse at that than reddit readers.

reddit readers spent more time on the page than digg users; digg users average stay was short enough one has to believe few of them actually read the entire piece.

I got some really interesting thoughts and feedback from reddit users. When the digg users arrived, I ended up shutting off comments. I may not have agreed with the reddit users -- but they were trying to hold a conversation. The digg users were posturing, and I found their feedback mostly annoying or arrogant, except for the abusive ones (kids, I was having flamewars before you were potty trained... you do not amuse me)

My bottom line:

Well -- I'd been experimenting with a "digg this" feature on typepad. It's gone. I see no reason to encourage those folks to come visit my blog. They generated a LOT of page views, but most of the constructive things I saw from the viewing tsunami I can attribute to reddit. Digg definitely bumped my page views, but didn't generate traffic elsewhere on the site and (to the best of my interpretation) limited subscription increases -- and definitely not increases to warrant putting up with all that other stuff.

Reddit -- different beast. Not as big a raw pageview bounce, but better and more thoughtful feedback, more people actually checked out the site (not just the posting) and more new subscribers.

I couldn't really track which links on other sites came from which original linkage, but if you compare the tone of the comments and the tone of the blogs that ended up linking to me, I could make some intelligent guessing. And if you compare those blogs with their technorati rankings (just as a raw measure of influence and visibility), well, the blogs from diggers that linked to me aren't ones that will generate significant traffic. Which kinda makes sense -- digg is where they seem to do their linking and commenting.

Being linked by digg seems to primarily benefit digg. Being linked by reddit seems more synergistic -- both sites benefit. I expect the subscription bump fades with repeated links, but there's definitely one there.

I've decided I can live without digg (but that's no suprise to people who know me or read this blog; this just reinforces with a little data what I already believed). Reddit seems to be a different crowd, a little less, well, full of itself. Having said that, getting linked does help drive new traffic in, but expect all but a tiny percentage of it to be transient.

Given that -- given that 99% of those readers will ONLY see that one page, I think it's crucial that those pages be designed to encourage folks to stay and explore and subscribe. Standard blog designs do a lousy job of that -- but mostly because I'm not sure that's anything we've ever really considered. It seems that having subscription links, recent postings and related posting blocks in prominent but non-annoying locations would be a really smart thing.

I do NOT think that simply putting them in the sidebar is what you want, in fact, I'm beginning to think that the individual article page needs a much different design than front page or index or category pages. Must think about that...

All in all, I'd define sites like technorati and reddit as symbiotic -- both sides benefit -- while digg looks more parasitic to me. These kind of bumps definitely have a value, but I think a blog needs a good design to help take advantage of the bump; the reality is, little of the bump is going to stick. I, personally, am not going to "promote" (hell, I almost said "whore") myself to get links from these sites, but when they happen, it's a good thing, if only an indication you said something people notice. I'll take that...