About a month ago (that’s forever in internet years) Marco posted the note Google and blogs: “Shit.” where the general thought was that his blog traffic was flat and starting to decline over the last year or so. There are a lot of reasons for this and Marco does a good job starting to explain it, but you can also point to the death of Google Reader as a big factor.
At the time this got posted I got into a conversation with Glenn Fleishman about all of this and noted that while my readership levels have to a good degree stagnated, I’ve also seen a significant improvement in engaged readership and more importantly (well, to me), a huge improvement in incoming revenue. Glenn encouraged me to write about what I’ve been doing which, of course, I never found time to do (sorry!).
Now Glenn’s posted a bit of a follow-on to all of this over on Six Colors (my favorite of the “post Macworld era” sites that sprung up after that Diaspora, although iMore comes close) talking about the Post Blog Era and how newsletters and podcasts are going to take over the blog space.
I think Glenn is partially correct, but partially mixing up a few things, so I think now is a good time to explain my view and hopefully clarify and not muddle the situation. I don’t think blogs are dying — I’m not even convinced they’ve hit “peak blog” as Glenn has called it, but I do believe we’re seeing some significant shifts in the expectations of readers and those of us who create content need to understand and adapt to those.
Over the last year — almost 18 months — I have been revamping my web presence to update it to my view of web sites like this should look and operate. In that time, my overall pageviews have dropped by about 10%, but my Amazon affiliate advertising, which is the only revenue generating aspect of the site, has gone from about $2/month to averaging almost $110/month. That means it’s now paying my hosting bills for my sites, which is a huge improvement for me. That number is consistent even if I get busy and can’t post new material reliably.
This is one of those places where I think Marco and Glenn are seeing the shift slightly sideways. it’s not that blogging has his “peak blog” and is fading, it’s that the era of the high-volume high-churn dated diary style of blogging has. What’s happened instead is that we’ve been seeing this big upsurge of interest in long-form content again — just look at Medium and what it’s accomplishing and trying to accomplish as one example.
This isn’t the point of “peak blog”, it’s the return of well-written and detailed (and/or thoughtful) content being favored over churn bait. About 80% of my page views on a given day are driven by organic search, and 99% of those are from Google. Organic search isn’t dead, but it’s been tweaked to give preference to (I think) longer pieces and to pieces that are better received by the readers — my guess is a factor they’re considering more strongly is how long someone is spending viewing a page. 10,000 pageviews where the average length of stay is 15 seconds (the “digg” model) used to be what people were chasing. Today, I think what the search engines are looking for is material where the readers are staying longer and clicking through off the pages.
This experimentation led to three big insights I’ve been using to guide my reworking of my sites:
The first big insight: Long Content wins
My most popular page on my sites is my Fuji page. It started out as about ten separate blog posts, which quickly got out of sync with reality. One insight I took when analyzing how users were viewing things was that multi-part series and links to updates in the blog failed miserably: people rarely if ever followed the links. They showed up, they looked at what was there and they rang off. Various experiments to try to encourage them to stay and browse or to explore related content mostly failed miserably.
In evaluating analytics on my site, I noticed that one specific piece that I’d long back rewritten as a single longer page (vs a series of dated blog posts) was being fondly thought of by Google and had significantly longer average reading time. That led to my first big insight: If I couldn’t convince people to explore around the site, STOP TRYING.
So I did. I use the blog to create content and make it easy for interested people to see new content and keep up with what I’m doing as a journal-style commentary, but the reality is that a blog entry sees 90% of its views in the first ten days and then it sinks into oblivion. even if one part of a multi-part series catches longer-term interest, it doesn’t drive much to the rest of the series. So the simple answer is: pull it all together onto a single page, polish that page, keep that page updated over time, and make it easy for people (and search engines) to find that content.
So when I updated my site, I threw out a lot of crap content generated in the “you must update daily or you’ll be forgotten” era, dropping the site from > 1,100 entries to under 400. Over a couple of months I took the best of the content and compiled it into a few key topic pages, rewrote the material so it worked together well, updated it as needed, and then deleted all of the original blog entries and added redirects to the new topic page.
The end result? Where any one of ten blog pages (all of them cross-linked to each other as part of a multi-piece series) might have seen a few hits, and if I was lucky 5% of the viewers would check out one extra page and be gone in under a minute, these new pages have staying power: the average time spent on these pages is over six minutes and the longest average visit is over ten. Ten minutes. One page. I have found that visitors to these pages, after having spent six or seven minutes on this page, are a lot more likely to do some poking around the site — front page, about page.
I think this is one reason why sites like Daring Fireball work as well as they do. They aren’t trying to churn you into visiting constantly, but when they do post content, it tends to be content that you settle into and read. It has depth and complexity to it, and it’s not there to simply try to get your eyeballs around some piece of advertising that pays mili-pennies per view or click. And that was my second insight…
The Second Big Insight: Users are blind to advertising.
I’ve had this long-term, experiment I’ve been playing with I call For Your Consideration. Back in the 1980’s I published a science fiction fanzine called OtherRealms which was, at its core, a review-zine. I also spent some time as Amazing Stories SF/Fantasy book reviewer back in the days when TSR owned the magazine. Reviewing is a writing form I enjoy, and I’ve been exploring ways to make it worth my time to do more of it, or even turn it into a revenue stream — with mostly negative results.
One of the things I did as part of those experimentation was throw out the standard Amazon Affiliate widgets — because they are frankly butt-ugly — and instead tried building some affiliate advertising blocks that were designed to look good within the design of the site. In other words, part of the content rather than grafted on generic widgets.
Much to my surprise, the clickthrough rate of my customized affiliate blocks skyrocketed, and so did buying through them. That’s when I realized that people quickly tune out the bits of a web site that scream ADVERTISING even if that advertising is relevant to them. So much of the web today is festooned with increasingly invasive, obnoxious, irrelevant and bluntly ugly advertising that most users simply stop seeing it. They’ve become conditioned to ignore it.
I tried a couple of tests. I took a couple of my long-form pages and put Amazon ads on it. I took a couple of others and put the affiliate blocks with my custom design on them. In one case I put the custom blocks with affiliate items that weren’t related to the content of the page.
The results were clear: the standard Amazon widgets did very poorly. The custom blocks with non-related items did somewhat better but not great. And the custom blocks with items directly related to the content saw very strong responses. The custom blocks don’t hide that they’re amazon affiliate links, I try to keep the disclosure open, and of course, once you get to Amazon the situation is obvious.
But the result to me was crystal clear: people’s web experience has been so abused by aggressive advertising that they’ve been conditioned to avoid it, even if the material is relevant and of interest to them. I call it being “snow blind” to the advertising. So if you want a good response, you have to keep it low-key, you have to keep it relevant, and you have to make it part of the content itself — without hiding it’s relationship to what it is, which is affiliate advertising.
From what I’ve seen, if you do that, people are cool with it and will follow the links and some percentage of them will use them. But even something as relatively un-intrusive (if butt ugly) as an Amazon widget box is enough to keep people from clicking over and using the affiliate link.
Insight the third: Social is a conversation, not a push channel
So, when Google killed off Google Reader, I figured that a lot of people following my site via RSS would disappear. I believe that in fact happened, and I probably lost about 50%. The other 50% seem to still be hiding out on my RSS feed using other tools, whether personal readers, sites like NewBlur, or whatever. I don’t do RSS analytics so I’m guessing, but I can interpret things via indirect means like watching image views on posted material, etc.
I felt a lot of the subscription population would replace RSS with social, especially Twitter, and so I decided to try to do it as well so I could understand whether and how well it worked. In fact, I was able to shift about 95% of my feed following to twitter just fine, with small outposts of material that showed up primarily on Facebook or Tumblr or Google+. I do use Newsblur for a few feeds today for convenience, but I’m primarily following people by their social presence.
I’ve done some experimentation with what works best for my sites, and I’ve found a couple of interesting insights. One is that it doesn’t really matter WHEN I publish a blog post, where in the RSS days things posted early in the day and on weekdays did a lot better than late afternoon or evening or on weekends. What does matter though, is that the social stream notices of content need to be there when a person goes looking for them. Generally not a problem on Google+ or Facebook or Linkedin (to a degree) because those sites seem to do a decent job of flowing new material to users (except when it doesn’t, and Facebook is notoriiously unreliable about being reliable about what it does, so you never know when it’ll bury thing you send or expect to see). Twitter, it turns out, has a half-life for a tweet around 4-6 hours, so if I post a link to a new blog items in the morning and you show up three timezones over in the afternoon, chances are you’ll never see it.
So with Twitter, I now use Hootsuite to schedule in repeating posts on new content; typically 4-5 times 4-6 hours apart so there’s about 30 hours of coverage. I’ll typically make sure material posted on a weekend gets some coverage on monday as well. While I typically post things early in the morning (pacific time) and then manually schedule in the extra tweets (because I’m too lazy to build a script to do it yet), it’s often the third tweet, which I normally schedule early afternoon pacific and early early eastern time that sees the biggest response.
I think it’s crucial that you don’t be too automated on social channels, especially on twitter. I know too many people who have little tolerance for what is clearly robots blaring out loudspeakers onto twitter. I try to make sure that when I do multiple tweets they’re far enough apart to not overlap in a typical user’s feed but instead try to cover the range of geographic time zone regions my readers live in: Pacific time, Eastern time, EMEA, and Asia/Australia.
That coverage seems to work best, and I get almost zero complaints. when I experimented with more frequent postings, I got noodged, so I backed off again. If I went less frequently, overall readership dropped off, so this seems to be optimal for making sure people see what I’m doing without pissing them off by blatting stuff at them too often. So that’s what I do.
What’s all this mean?
Here’s my view of life in the content world today:
- Stop worrying about pageviews. it’s a stupid stat. Worry about engagement. I’m a lot more interested in average time on a page and whether they click to other pages (or affiliate links) than I am how many people hit a page.
- Long content rocks. Use the journalistic aspect of a blog to generate the content over time, but make sure you pull it all together into fewer, longer, well-written topic specific pages over time. If you can do that IN the blog, okay. but I think the model of “short blog posts” -> “longer rewritten blog page” -> “ebook of pages” is the model most of us should be thinking about (and yes, ebooks is the next phase of life for that content, whether given away as part of other promotions or moved to Kindle/iBook or some form of paid published content. if nothing else, thinking like this will help you focus on writing quality content and away from the kind of “blog filler” crap that wastes time and generates no real value…
- Users are blind to advertising. So stop wasting time and page real estate on advertising that pisses off your users and doesn’t really work well anyway. If your business model is dependent on that, start building a new business model, because payback rates for that stuff is only going to get worse.
- But users aren’t against low-key stuff like affiliate advertising, if it’s done well, compatible with the site and relevant to the content their viewing.
- Make it easy for people to find other stuff; most won’t, but some will, and you want to encourage them to like you. You don’t need hundreds of pages of stuff — pick out your ten best pieces, polish the hell out of them and focus your visitors on them. Then try to make the rest of your content that good and interesting.
- You really are better off with a dozen really good pages than a hundred mediocre ones. Focus on good, not lots. That is the exact opposite of what people have been telling bloggers for years, but the most valuable content on your site is the content that’s valued by organic search, and that content is the longer, more detailed and thoughtfully written pieces, not the chatter.
The funny thing is, that’s pretty much what Google’s been telling us all along. Good content wins, but you aren’t going to generate good content at 400 word chunks daily — unless you pull it all together and build it into a few longer, better pages.
Where I’m headed
I’ve been in an experimentation/study and redesign phase on and off for about 18 months. For the last couple I’ve been implementing some plans that came out of it, and part of that was to split the photography content off to its own site, and now I’m mired in a redesign of this site — delayed because I’ve been working on “my real job” projects instead.
But the goal is to implement the above. I made the decision to split out the photography content onto its own site for a couple of reasons:
- It’s now mature enough with enough of an audience to stand alone
- It lets me add on new photography-related projects on the side more easily.
- It frees me to open up this site to explore new content areas again, giving me potentially new content areas I can build out the way I did the photography stuff.
I’m in process of doing the edit down and focus on this site the way I did the photography. It’s still very much a work in progress, but it’s going to involve more of the review material, more tech industry talk, and a return to my sports writing on a more regular basis. I have no idea which, if any, may build into a big enough area to warrant it’s own site, but we’ll see what happens.
Over on the photography side, I’m working on a podcast to reboot the Before and After series I was doing for a while, and there will be a Patreon aspect to that (if for no other reason than it’s time to start experimenting with that model). More on this soon, part of my time when I went and hid in Fort Bragg was giving myself some focus time to do some planning on this project.
And I agree with Glenn that both Podcasts and Newsletters have a key role to play in content distribution moving forward, but I see this less about replacing blogs as about bringing forward content into the way people are starting to consume it. Text will continue to be a prominent piece of that equation, but Podcasts are content for times when text is inconvenient or impossible, especially commute and exercise time. And video (i.e. Youtube and/or Vimeo and/or etc) is increasingly about casual research and browsing — tablet couch time — and isn’t so much about replacing text as about making it possible to cover material for which text is a lousy way to deal with it. Such as my before and after series, where I found doing it as text mostly failed, but where I think a video component backed up by a textual part will make it a persuasive and interesting series.
Online video is both content sensitive (some things just need video to be interesting) and generational (younger people are conditioned to find it on Youtube and expect it to be in video form). This and audio podcasting such as what Jason and Relay.FM are doing don’t replace blogging so much as open up areas that are increasingly important to users given the technology they have and who are looking for content to fill those times in that form.
Ultimately, it’ll all work together and if done well, leverage each other.
At least, that’s my hope.. And where I’m headed…