Internet Marketing Blog

Separate The Blog Wheat From The Chaff With Comments Heat Maps

How To Evolve Blog Commenting In Three Easy Steps

I've been thinking about social media a lot lately  and how people interact with Web 2.0 technologies and services themselves as well the other people who use those services.

In particular, I've been thinking about the comment function of blogs. And how to improve that function.

As a football nut and a die-hard Vikings fan who still reads everything about the team despite their abysmal record this year, I spend a lot of time on Vikings blogs, especially the Star Tribune's superb Access Vikings blog by their beat writers Judd Zulgad and Kevin Seifert. 

I'm sure it's one, if not the, most popular Strib blog. There is certainly a lot of participation by its  readers. A post from yesterday about Daunte Culpepper coming back to town boasts 162 comments.

I often read the comments because there's some very good stuff in there, but wading through the crap and the banter from one reader to another is usually a waste of time. Separating the conversational wheat from the chaff is a problem for all blogs with a lot of reader participation, not just Access Vikings.

Step One

I'd love to see some smart programmers at Six Apart or Blogger or some open source guru create a plugin for WordPress that would do this: Allow me (as a registered user of the blog in question) to annotate the blog post and apply highlighting and/or my own comments in the form of, say, a sticky note(s) that I can place anywhere on the page.

Step Two

It would also have a function like the fantastic ClickTale service that gives you a "recording" of each site visitor's behavior on the page by following mouse movements. I often, for instance, click and drag my mouse over text to highlight interesting text. I don't know why I do it, but I do. That behavior could be captured.

Step Three

Combining ClickTale recordings with the sticky notes data should give you enough aggregate data to create a heat map for that page.

Apply heat maps to the blog comments and, voila, as a blog reader I've got a quick visual representation of the relative "interestingness" of blog comments. And as a blogger, I'd get extremely valuable data on how people were interacting with my blog.

Perhaps someone could come up with a semantic solution that analyzes the text of blog comments and the frequency with which blog comments refer to other comments or commenters to determine "interestingness."

However it's arrived at, we sorely need a solution and it seems to me that the technology is probably already in place to create it.

New e-strategy.com Launched

I've apparently got web site redesigns on my mind this week.

My first two posts this week were about the NFL.com redesign and now this one. As Kurt Vonnegut would say, So it goes.

It was long past time to overhaul e-strategy.com. This will be the fourth version of the site since I first launched it in December of 1998. The most recent version has been in place since October 2003, so, as I said, it was about time to redesign.

Back in the day, I once got a hair cut that was a completely different style than I'd ever had before. It just wasn't me. So until my hair grew out enough to restore it to the previous style, I walked around embarrassed and feeling like a dork.

That's sorta how I have been feeling about the last version of e-strategy.com for a few years but I've never gotten around to actually doing the redesigning. It was a fine looking site when it first launched but it was designed before the Firefox browser was released and the only browser you needed to design for was Microsoft's Internet Explorer (IE). The site, therefore, was designed for Internet Explorer and it looked just fine that way:

But Firefox gained marketshare to the point that I started using it, first for site design purposes, but quickly as my everyday browser. I could see why it was popular. It was fast, it had tabbed browsing, and it had all kinds of add-ons and extensions.

Last month 20% of the visitors to e-strategy.com used the Firefox browser and 29% used a browser that was not Internet Explorer. Internet Explorer still is far and away the most dominant browser, but the percentage of non-IE surfers is too significant to ignore.

I was so embarrassed by the site because this is how it looked in Firefox:

Ugh. The whole thing was just broken in Firefox.

Redesign Goals

As I said, a redesign was long overdue. I had five goals for the redesign:

1. Maintain some of the elements of the old version because I like them
2. Ensure the site looks good and works well in IE, Firefox, and the Safari browser
3. Avoid using HTML Tables for layout
4. Use Cascading Style Sheets for layout and presentation
5. Integrate my various Internet outposts into the site

I accomplished my firs two goals. The site maintains some of the look and feel of the previous version and it looks good and works well in all three browsers:

With the exception of the front page, the entire site uses CSS for the layout. Cascading Style Sheets has display quirks between the browsers and for that reason and to save a lot of time, I used tables for layout a little bit on the home page.

From a search engine marketing standpoint, the reason you want to use CSS instead of HTML Tables for layout is that it produces leaner code, and that requires much less work for search engines to read and understand your site. It also allows you to position the most important content where the search engines will read it first.

Finally, it reduces the code-to-text ratio; search engines do not like sites that are heavy on code compared to test as much as they do sites that are lean on code but have plenty of text.

Web 2.0 & Social Networking

The previous version of e-strategy.com was launched before the advent of Web 2.0 and the social network and sharing sites. But now e-strategy has this blog, a MySpace page, a YouTube channel, a Technorati profile, del.icio.us bookmarks, a Twitter blog, and a Flickr account, among others.

I wanted to pull all of those online outposts together to create an e-strategy network and create upstream traffic from each to each, so each site has been built into the navigation system of the site.

These blog posts are automatically updated on the home page via a FeedBurner feed and our YouTube videos are embedded into the site.

I'll be building the site out a bit in the future and I'll keep you up to date as I do.

NFL.com Redesign & Search Engine Marketing Blunder

Darren Sharper Tackles Cedric Benson 
Originally uploaded by NicelySighted

The fundamentals in football are how to block, how to tackle, how to catch and how to pass. One of the fundamentals of redesigning a web site is to preserve, or at least account for, existing inbound links to the site.

So it was odd to discover after the redesign of NFL.com was unveiled amid much fanfare, that they changed the URL structure of the pages of individual football player's profiles but failed to account for all the links that pointed to the old player profiles.

Pro Football Bloggers' Links To NFL.com

According to Google, there are 22,800 inbound links to the root NFL.com domain. As of this writing, Technorati lists more than 67,000 blog posts linking to NFL.com. That's a lot of links representing a lot of traffic and a major marketing blunder.

As someone who blogs quite a bit about pro football, the league's failure to account for existing links to their site is especially annoying because all of the links I have to player profiles prior to the redesign are now broken.  This is obviously lost traffic to NFL.com, but, more importantly to me, the broken links create a horrible user experience for the readers of my blog posts. They get is an error page rather than the player profile they were expecting.

The extremely frustrating thing for me is that long ago I made it policy to link to a player's NFL profile under the reasonable assumption that the links wouldn't change. The problem of linking to team profiles is that players change teams though trades, free agency, and cuts and that results in broken links. It's reasonable to assume that many other football bloggers came to the same conclusion.

Perhaps I should only link to Wikipedia player bios from now on.

But it is not just links from blogs that have been broken, the links from the search engines are broken, too, and that's a user experience and branding problem for the NFL.

NFL.com's Search Engine Optimization

I'll use Minnesota Vikings safety Darren Sharper to illustrate NFL.com's previous URL structure for player profiles. The URL for Sharper's profile on NFL.com was http://www.nfl.com/players/4249.htm. The new URL is http://www.nfl.com/players/darrensharper/profile?id=SHA500479. Note that they've included the player's name in the URL itself, while the old URL only used an ID number.

As of this writing, this is what the Google search results look like for "Darren Sharper":

You'll notice from the screenshot above that the link for Sharper's old profile reads "NFL.com #42 Darren Sharper." That text is taken from the HTML TITLE Tag of the page. The new player profile pages use only the player's name in the TITLE Tag. Couple that with the inclusion of the player's name in the URL and it becomes obvious that the redesigned site is intended to be search optimized in order to boost the page's rankings on the search engine results pages.

If the NFL knew enough to optimize the redesigned site for search engines, wouldn't they know enough to account for existing inbound links? Apparently not.

When you click on the link to Sharper's profile from the search results, you get an error page. That creates a frustrating user experience for NFL.com and that frustration hurts the NFL brand.

Site Redesign Fundamentals

The fundamentals for dealing with an issue that many site's must deal with are well known and relatively painless: 1) permanent 301 redirects to seamlessly point an old page to a new one in a "search engine friendly" manner so that you do not lose your search ranking, and 2) using an XML sitemap to tell the search engines about your new pages. NFL.com does neither.

The NFL certainly doesn't need search engines to drive traffic to NFL.com. They've got television to do that. But the fact remains that NFL.com is listed in the search engines and accounting for broken links should have been done with the redesign. It's obvious the NFL has a pretty sizable Internet marketing budget, so you'd think their Internet marketing team would have planned for that.

Blogger Liaisons

Any organization with mass popular appeal that is likely to have a lot of people blogging about them--which is especially applicable to professional sports leagues--should think seriously of creating a blogger liaison. This person would have a blog themselves and reach out to bloggers, solicit their feedback, inform them of issues that may affect them and answer questions, etc.

An obvious example that will immediately spring to mind for those in the search marketing industry, is Matt Cutts who performs that very function for Google's relations with the search engine marketing community.

If the NFL had a blogger liaison and informed pro football bloggers about an upcoming site redesign, I'm absolutely certain those bloggers would have worried about broken links and the NFL would have become aware of the issue prior to the launch of their redesigned site.

e-strategyblog.com vs. estrategyblog.com

Case Study: Domain Name Strategy

Hyphens in domains are problematic when it comes to earned media, online or off.  A perfect case in point presents itself to me today.

As I posted minutes ago, yesterday I gave a radio interview to a reporter from Minnesota Public Radio for the Future Tense program about citizen journalists' role covering the Minneapolis bridge collapse.

At the end of the interview, the reporter cited this blog but omitted the hyphen: "David Erickson runs  estrategyblog  dot com." I don't fault the reporter; it is entirely too common for people to omit hyphens from domain names when they cite them. It's just an inherent danger that you have to live with when you use hyphenated domains.

I bought e-strategyblog.com to match my company's domain: e-strategy.com. But as a result I have to emphasize the hyphen when citing the address: e dash strategy dot com, e dash strategyblog dot com.

People usually go to search engines if they want to find out more about something they discovered offline. So, in this case, if they wanted to read this blog, they'd either search for my name or the blog's name as they heard it: estrategyblog.com.

If I don't already have that domain in place, then it's likely people will be frustrated when they try and find me.

And I did not have the domain in place, so I'm very likely losing a lot of potential readers. And the Future Tense audience, because it is technology focused, is very likely to like this blog.

Here's how I'm addressing the situation:

1) I bought the estrategblog.com domain and I'm having it forwarded here (as of this writing, the forwarding has yet to take hold).

2) I'm writing this blog post and seeding it with the keyword estrategyblog.com in the hopes of it showing up quickly in search engine results (and blogs do tend to show up faster than web sites).

If you're planning a new web site or domain take these things under consideration. If you want a hyphenated domain, chose one that also has a non-hyphenated counterpart and put the forwarding in place.

See also:

• Domain Name Strategy - Follow Up

Communism 2.0

Since I've been going on about the open source movement this week, I thought I'd republish for you a piece I did originally for the May 25, 207 issue of the Politics In Minnesota newsletter:

It's often amusingly accurate: The type of technology you favor often predicts your ideological leanings. Mac users, for example, are passionate about their computers and can often be heard railing against Microsoft, the evil megacorporation; the irony of their passion for Apple, another megacorporation, apparently lost on them. The same phenomenon can often be seen playing out with Internet technologies, as well.

There are two primary technologies used to create complex database-driven web sites and application: Microsoft's ASP and the open source PHP. Both programming languages are fine and will do the job. The primary difference between them comes down to one basic thing: cost. Microsoft technology requires licenses and software purchases and the software required for PHP sites is free.

Free versus paid. Open source versus proprietary. This may explain why tech people who lean conservative are more likely to use Microsoft technologies while the more liberal tech people tend to favor open source; and it plays into all the political stereotypes.

Those who favor Microsoft technology are more likely to come from the corporate environment and probably have more money to pay for software. Those who favor PHP, on the other hand, are more likely to come from a nonprofit environment or are freelancers for whom the savings on software are significant.

Proprietary software such as Microsoft's is more top-down because it prohibits the user from modifying or enhancing the source code. And the purpose of Microsoft software, of course, is first and foremost to make money. Open source software, on the other hand, is about collaborating and sharing and creating a pool of resources (i.e. property) which anyone can use and/or improve in order to enrich the commons rather than the individual.

The economic stereotypes fall into place.

On Monday, the Washington Post ran a story saying that Republicans are well behind Democrats in online politics. That may be explained by the aforementioned embrace of the respective technologies both in fact and philosophy.

The open source movement has gained tremendous momentum largely due to the fact that open programming platforms such as PHP and JavaScript have powered Web 2.0, the Internet as it looks today. Liberal tech people therefore have more experience, skills, and expertise using the current lingua franca of the Web.

Further, it is the notion of open source (where anyone can contribute) that informs much of what is driving Internet content today. Wikipedia is a massive online encyclopedia to which anyone can contribute and/or edit. Blogs (and, increasingly, newspapers) allow anyone to comment and hold a discussion pertaining to a given blog post. User-created content seeds such sites as YouTube and Flickr. And social networking sites such as MySpace and Facebook and Twitter, make it easy to find and communicate with people of similar interests.

Such a chaotic environment may feel comfortable to debate-loving liberals but not so much to message-disciplined conservatives.

As if to confirm this ideological breakdown, the last remaining prominent Communist, Cuba's own Fidel Castro, has encouraged his country to adopt open source technology while Brazil's Lula Da Silva began migrating government computers to the open source Linux operating system in 2004.

Open Source Vs. Proprietary?

I've recently been looking into a lot of open source web-based technology and I've become more and more impressed with the solutions. Most of the Web 2.0 technology is built using open source software or platforms of one type or another.  Open source solutions can be particularly appealing to small businesses and nonprofits, primarily because they are free.

There are basically  three types of open source products: 1) Actual software like Linux or OpenOffice, 2)  scripting languages like PHP, or web services or applications like WordPress.  Despite my previously discussed gripe about the lack of documentation, I can give you several reasons to consider using open source:

1. It's free.
2. The premise of open source is that by taking advantage of a global community of developers, and because the code is available to anyone, you can more quickly create stable software and more quickly identify problems/bugs/glitches and more quickly fix them. And, because you have a global pool of talent that is contributing plug-ins and extensions back to a given project, the features and functions are expanded a great deal at the same great price.
3. The open source Web platforms I've looked at and/or worked with a extremely sophisticated and in most cases, match commercial products feature for feature.
4. Open source just looks like the future to me. Who would've thought, but it is pure communism! If people are willing to give such sophisticated solutions away for free, businesses are going to adopt them and there's no way commercial products will be able to compete. I'm not suggesting you switch over without a thorough consideration of your needs and capabilities, but there's a reason Google uses the Linux operating system.
5. In many cases, open source offers you an out-of-the-box solution so it makes no sense to try and build one from scratch when you've got one already available for free.
6. I've been increasingly seeing more and better open source Web platform solutions written in PHP than stuff written in Microsoft's development languages like ASP.

Open Source Manual Writers

Open source and I did not start out on the right foot.

Open source is software that is developed globally via the Internet by anyone who wants to contribute through freely available source code. Developers can contribute by fixing bugs, adding features, or extending functionality with the understanding that the resulting software will be free for anyone to use and modify, if they want.

My first experience with open source was a project that required we use ChiliSoft, the open source version of Microsoft's Active Server Pages scripting language for building dynamic web sites.

It was a nightmare because there was no documentation you could consult when you ran into a problem. If you encountered an error or bug, you had to just sorta figure it out yourself. Microsoft, by comparison, has extensive documentation for ASP.  If you encounter a problem, there's a place for you to find a definitive answer to solve your problem.

While I've warmed considerably to open source solutions, the lack of documentation is still a major drawback if you decide to use open source. If there is a user manual included, it's most likely bare bones.

What generally passes for documentation for a given open source project is an online discussion board where you can find threads that typically begin with "Has anyone had this problem...?" followed by a discussion of the board members trying to diagnose the problem and sometimes arriving at a solution, sometimes not.

The open source movement has a lot going for it: Full featured, sophisticated, extensible, and stable software solutions that, by the way, are free. But the major flaw is that because of the lack of adequate documentation, you end up spending a lot more labor diagnosing and solving problems than you would with proprietary software.

What the open source movement desperately needs are open source writers. Not coders. Coders are usually not experts at the written word and, because they're technically proficient, they often take for granted that the people reading the manuals will have the same technical acumen as themselves. When coders write manuals, the manuals are often unreadable.

What the open source movement needs are people who can write software manuals in lay language. You occasionally see a wiki  that's been set up to serve as the documentation for an open source project.

That's a great idea, but the wiki's I've seen used for this purpose have been hit and miss, often incomplete and inadequate. It seems to me that there should be some agreed-upon standard for open source documentation that people working on open source projects can come to expect.

Perhaps the solution is to include within the message board software prompts to encourage users to contribute their solutions to the wiki manual.

I just don't think that open source software can really gain mainstream adoption until this particular problem is solved.

Apple's Safari Web Browser For Windows

When I read yesterday that Apple had released a Windows version of their Safari browser, I rolled my eyes. The last thing I need to worry about his how clients' web sites display in a browser that uses a fundamentally different rendering engine than the two dominant browsers, Microsoft Internet Explorer and Mozilla Firefox.

I've worked on projects that required a "Safari compatible" web site and I gotta tell you, it's no fun at all. Since the browser displays web sites often radically differently than Internet Explorer and Firefox, you end up spending a massive amount of time testing and refining the code to display properly in all browsers. It is just a phenomenal amount of work for such a tiny audience.

So I downloaded Safari and tried it out and so far so good but I've only used it on a few sites. Despite what Steve Jobs says, it just ain't all that fast. It hasn't crashed on me yet but I am apparently the exception.

Safari will apparently be the development platform for the iPhone; if that device takes off, then, Safari testing may become routine for anyone doing mobile web marketing.

Ugh.