My Ideas on Multiblogging

Traveling Detail Light

When studying a new technology, I plunge in headfirst. I have read a lot of books and from experience, despite what I learned from books, I still had to have a picture of what I was trying to learn on my own.

Up to that point, I was basically following directions, hoping that the steps would give me a clue as to the reasons behind the process. Because once you have the reasons things are done and the reasons why they are done the way they are, you can travel lightly.

When studying SEO, I learned a lot of techniques and read a lot of books. I had to, until I could boil it down into principles that meant something to me. Get links to your site and use good keywords. That's basically all I carry around now. From this, I can infer all the techniques I need to, on the fly. A bigger and more detailed description of SEO has to be updated more frequently in a volatile environment like the Internet. Much better to keep it simple and let the project determine the details.

Rinse and Repeat

I have wanted to create a network of sites ever since I built my first one. I built my first one in my head while I was sitting in jail for my first and only DUI. I have about 35 domains, not much for a domaineer but a lot for someone planning to build a network of sites that he would run.

But I would doing it by slowly automating and outsourcing things whenever I could. The only actual part that would require my attention in the Utopian ideal is writing and interacting with other people. Everything else could be farmed out or handled by software. Or at least software could bring all the diverse tasks from all the diverse locations batched together by type into one place.

But my desire to build sites overtook my desire to have an infrastructure to support said sites. A task that takes 5 minutes per site adds up when you add more sites to the mix. Soon I had all the sites I could handle and then some that just sat there for months at a time.

Multiple Launchpads

Whether or not that was an actual mistake or or not is what I have to ask myself. Because I did it with the idea that I needed to spread my efforts out. Having a domain was worthless without doing something with it. My domain buying got ahead of me, but I did get some work done. Here are domains I own in each Pagerank level. I had a PR5 domain for about a year. When it went back to 4 the next year, it was expected. I could not for the life of me figure out how I got it.

  • PR4: 3 sites
  • PR3: 6 sites
  • PR2: 4 sites

There were a lot of PR1's but I got tired of looking. Yes, I don't actually have all my sites bookmarked. That would probably help. And for those Pagerank detractors out there, it does indicate the possibility of ranking for a phrase against competition.

Time To Do Something

It came down to two pieces of software that I would do this on, Wordpress or Drupal. I played with Drupal for a while, found it stronger for building very specific sites, but the amount of developers building Wordpress plugins meant I may have some of my programming done while I wait. I still have uses for Drupal and have a few sites where it is one of the only solutions.

So I stuck with Wordpress, built a theme, wrote a plugin and got used to using it to see if it was something I wanted to stick with. This was when I was still using a blog as an easy way to add content to a site. Many of the sites I built didn't have a date based archive.

But I got caught up in this a while and stopped the chase for the solution. But at the same time I was searching for the solution to high-profit streamlined monetization. The two had to kind of meet for anything to work. I had gone the Adsense route and the affiliate route. If the first was paved in gold as some claim, then the last was paved in what could only be diamonds.

The Ideal So Far

I say so far, because I know changes will be made as the reality of doing comes into play. But for now, it's good to have an ideal to shoot for. I use "algorithm" instead of "process" and I just mean a basic yes/no flowchart to use when evaluating something not software.

First there are tasks that can easily get batched together by the types which are:

  • Answering emails
  • Moderating comments
  • Answering comments

Then there are those that are best batched by site or project because it would involve too much switching gears to batch them by type:

  • Research
  • Writing posts
  • Visiting other blogs in that specific niche
  • Promoting visibility in that niche

It would be good to have an idea of the lifespan of a blog set up in the network.

  • Research the niche
    • Affiliate products
    • Competition
    • Find social neighborhoods
  • Buy the domain
  • Quick installation
    • Theme already chosen
    • Fantastico or Wordpress MU
    • A default set of plugins waiting to be uploaded
    • A script that populates the initial configuration of the plugins.
    • The default widgets are set up in the sidebars
    • Set up categories based on prior research
  • Prewritten flagship articles waiting to be timestamped covering a month of posts
  • A Static Promotion Algorithm for the first month
    • Submit to feed directories
    • Submit to site directories
    • Any potential for niche software to give away?
    • Any potential for free reports to give away?
    • Anything that can be released to the press?
  • An upkeep algorithm
    • Posting (answering comments batched with other blogs)
    • Commenting
    • Social Networks
  • An old site never dies
    • Get sold to the highest bidder
    • Collects enough Adsense clicks to pay for it's $8 a year plus MacDonald's
    • Increases in authority with age.

Basically all the static promotion could be outsourced in various ways by either using software or hiring freelancers. I spend $2000 a month on Adwords currently. Dropping $200 one time to get a site off the ground should be nothing and cut a lot of time.

The system could be made to start off initially without any PPC until ranking in the search engine is achieved and long tail keywords determined. Then using Adwords will become icing on the cake

The upkeep algorithm is shaky because everyone has different opinions on how often you should update a blog. I am not really sure either. That one will have to be tested with the niche with a minimum of a post a week.

The upkeep algorithm might also involve profile sets. Sets of profiles for different social sites and sets of blogs in that niche. If I work on one blog in the network, I want the neighborhoods associated with that blog close at hand.

As far as how often to start a new blog goes, I'll let the work load determine. A certain time should be spent on upkeep each week and some time for new projects.

How many can I run like this at once? How about guest bloggers? How about group blogging? How about paying for articles? Questions yet unanswered and added to research for the future.

The Solution to Batching

My initial idea was doing everything from one control panel. But then I thought better of it. A better solution would be not having the need to visit any Wordpress installation as often. The initial setup up should take care of the visits.

I write most of my post in Windows Live Writer. That can handle my posting solutions. I like to use simple tags and the HeadSpaces2 plugin, so when I post, I choose the "Post Draft and Edit Online" option. It would be cool if the drafts could be grouped by blog though instead of thrown in the same folder.

Comments could be handled at one time via email. As far as email goes, I have been thinking of switching from Thunderbird to Gmail, having everything come into one account and filter them in folders there.

If you have all of your blogs on the same server, you can use Wordpress MU with the multisite manager plugin and clone sites on the fly. Sites will use plugins from the same folder. One Wordpress upgrade upgrades them all. But all of these sites would be coming from the same IP which may not matter too much if you aren't too aggressive with your SEO work. I run some plugin heavy installations and a way of just moving the plugin folder to a new installation would be great.

To do this across domains, I could build a generic Wordpress installation on my laptop where all plugins are up to date. I could have my widgets set up and everything. I could then just upload it to the new site and change details. If upgrading is as simple as uploading the new files and upgrading the database, then that could be done that way also.

A cron script on the local Wordpress installation could check and initiate Wordpress and plugin update checks and download the new plugins. This would be logged and the logs checked for update news. Then an update script or macro would do it's job automating on each blog in the network after I approve it.

Current Research

But the ultimate key to batching would be having both stats, monetization, and a central index of your posts in one place. They would be three points of the triangle.

It is logical that some of your posts will overlap, so let them via linking with a blog network version of alinks. Or the central index of feeds could be searched for matching tags and related posts listed in the sidebar

A keyword you have gotten a hit for is worth more than one you haven't. Use actual keywords inside the alinks installation on the fly. A hit anywhere on the network instantly triggers links across the network.

If the page is monetized, this should be known centrally and then potential PPC campaigns could be created with the same known keywords. Yes, even if you are #1 for a term, there are still people who click the ads first. The adwords data should be in the mix so that you can see the ROI by just navigating to a page.

As far as monetization goes, this would work great for Adsense, but I'm going for affiliate products mainly. Details on that to come.

This post was heavily inspired by a post on Dosh Dosh, which gave me the initial oomph to create a blogging network blueprint tailored to me months ago. It took this long to get part of the picture.

And when I got to the end, there seem to be more questions then answers. And the answers that did come were more like potential solutions than anything.

More to come.