Ever since I discovered the rel=nofollow attribute, which prevents Google from counting a link as significant as far as page rank is concerned, I have had mixed feelings about it. At first, I though it was a good thing, assuming that it would improve commenting quality and reduce spam. I have since changed my mind.

I have come to the conclusion that rel=nofollow is only good up to a point, and have removed this attribute on this blog, for those commenters who post at least three quality comments. This encourages people to come back and leave more than one comment, and gives me the right to ruthlessly delete any comment that is not quality.

For this reason, I have posted a policy on commenting on this site. If you leave comments which don’t appear or are deleted, reading this will give you a clue as to why it didn’t make the cut. I don’t plan to be ultra-strict or make it hard to comment, but I do expect good comments in return for commenters getting a page-rank-boosting link back to their site.

This allows me to be consistent with my initial opinion about rel=nofollow, and with my decision to promote blogsphere with meaningful links from comments.

In making this decision, I was influenced by posts on several other blogs, including JLH Design Blog, Sabastian’s blog, More Earnings via Search Engine Optimization, and weblog tools collection.

Which nofollow Eliminating Plugin?

Having decided that rel=nofollow is history on this blog, the question became: which plugin?

Via a short post on weblog tools collection, I discovered a succinct and useful roundup of plugins to remove rel=nofollow. Based Andy Beard’s list, I chose to use Link Love.

Link Love removes the rel=nofollow from commenters after they have made a specified number of comments. The number is rather high: 10 comments, but you can change this. I will explain how to adjust the number of comments in Link Love in a separate post, but the short version is that you have to change something in the code. This is easy to do however, and you don’t have to be a programmer to change this value. I have changed it to 3, in other words, if you leave three quality comments on this blog, links from your comments count for something as far as Google is concerned. Ten is high for all but the most popular and heavily commented blogs, I would think.

You may wish to operate differently. Read Andy’s roundup of plugins to remove rel=nofollow and decide which plugin works best for you and your site.

Links to other articles on this subject:

• WordPress (a little background, and links to pro- and con- articles on the subject, as well as various plugins).
• The Marketing Technology Blog

DoFollow plugins for WordPress:

• DoFollow - a plugin that removes rel=nofollow from URLs in comments. “Optionally you can also set a comment age limit for adding the attributes.”
• WordPress Plugin Database - links to several dofollow plugins - and many other WordPress plugins!
• Andy Beard - lists several plugins with a brief description of each. Highly recommended.

I recently set up an ecommerce website using WordPress and Instinct Software’s ecommerce plugin. I have already written briefly about customizing WP Shopping Cart for WordPress. I explained how I had to hack the system to get the look that I wanted.

I a simple suggestion for improving how WP Shopping Cart outputs its HTML so that the process of customizing the look becomes much easier.

Continue reading

I recently wrote a post about setting up an ecommerce system for digital downloads, specificaly my frustrations with finding an ideal platform, and my choice of wp shopping cart for WordPress as the closest fit, even if it really is not ideal.

The good news is that Dan at Instinct Software is monitoring comments about the software that he and his company have created, and are willing to do something about it (see comment on related article about ecommerce).

This post is for the benefit of anyone developing an ecommerce platform, especially Instinct Software who which to improve their existing system. Feel free to add comments if you are either a developer of ecommerce software for digital downloads, or if you want a better platform for selling digital download products.

Continue reading

A few things are beginning to come clear to me about how Google AdWords operate. It made little sense to me yesterday, but it is a little clearer to me today.

Firstly, there are some important differences between ads served up by the search engine (which appear to the right of search results) and the same ads served up on websites who operate AdSense.

Ads appearing along side Google search results start appearing pretty much as soon as the campaign is set up.

What seems to happen is that Google runs the campaign for a few thousand impressions, and determines how well the ad performs, in terms of how many times it is clicked. If not many people click on the ad, then the bid price goes up, even if there is no competition. The bid price seems to fluctuate according to the precise wording of the ad, the wording of the website (pay special attention to text in headings etc.), and how well the ad performs.

Continue reading

Just an hour or two after posting my previous article about setting up an ecommerce system, in which I mention the steep rise in the price of the bid for my key phrases, I have found they have dropped by a lot. One was going for $6.00 (which is really too high to be seriously considered as a competitive bid) has dropped to $1.20, while others that were set at $1.20 minimum have dropped to $0.23. I’ll try $0.23 for a while and see what happens.

Mean while, I have signed up for Yahoo Marketing to see how their service compares.

I signed up via a web page that offered $50.00 in credits, but I didn’t get the credits, so off to a bad start.

Worse, there are lots of advertisers using the same key words as I am using in Yahoo, where as I had the place to myself in Google. I find this strange. It may simply mean that Google advertising does not work well for the key words I am using.

As noted in a post of just a few hours ago about advertising my new ecommerce site, my first impression of using Yahoo to advertise my site was tainted when I failed to get the $50.00 credit I was promised. I should note that the site, payperclickdeals.com, is pretty positive towards Yahoo. I also got to their website via a paid ad in Yahoo search results.

Another thing I noticed: by default, everywhere outside North America is blocked. They don’t ask what regions you want to include (Google is very fine-grained about this), they automatically block Asia, Europe, Australia etc. and you have to unblock them. I find this unfriendly.

In general, I prefer Google’s user interface. Yahoo uses more style, Google has more features and a cleaner look.

The next thing I noticed is that it is a crowded market for my key phrase, which is hardly Yahoo’s fault. At first, typing in my key words did not bring up the ad, but about an hour and a half or two hours later, I did see my ad place in second position, which is pretty good, given the crowded market and my bid of $.30.

As far as I can tell from my server logs, I’ve not got any traffic from any Yahoo ads, but it is too early to be sure what that means.

I did notice though, that the statistics for ad impressions and clicks are not very current. The best they can do is display what happened yesterday. I think this is too much of a lag. Things change too quickly to make decisions for tomorrow based on yesterday, especially for a new campaign. I may change my mind on this once I’ve been in the game for a while. Google shows statistics that are two or three hours old.

I came across this article on BoingBoing, which is pretty negative about publishing Yahoo ads. Apparently it is up to the publisher to block visitors outside of North America from seeing ads on your website! Of course they took Yahoo off of their websites right away. Some attitude Yahoo has! No wonder no one suggested I advertise on my sites with Yahoo.

So I’m feeling slightly negative about Yahoo right now. But I have not yet come to any conclusion.

After a lot of work, I now have a complete end-to-end ecommerce system. Is it perfect? No, far from it. Each stage of the system can be improved quite a bit, but if I wait until everything is perfect, I’ll never get started. Firstly, no system is ever perfect. This is even more true on the internet where things change hour-by-hour, quite literally, as I will explain in a moment.

Let’s review what I consider to be a complete end-to-end system at present, noting that complete means the bare minimum to conduct business on the internet, not comprehensive with every imaginable feature one could throw at an on-line business .

Firstly, I have a product, in this case an ebook. Delivery involves a download .

Secondly, I have a website to sell it from. I’m still not revealing the website or the product, simply because I want to conduct and experiment that excludes search engine optimization .

Thirdly, I have a way of taking payments. To keep things easy in the beginning, I am using PayPal , and PayPal only. It accepts credit cards without the need to register.

Finally, I have a marketing method. To begin with, I am using Google AdWords exclusively.

Producing the product was a lot of work, but it was straightforward. I am using someone else’s content (legitimately and legally I might add), but I did have to put a great deal of effort in to turning it into an attractive pdf that is worth buying. This took a lot of work the first time, but it will be much less work the second time. Still, it is time consuming and takes certain skills and talents.

Developing the platform also took a fair bit of work, even though I used freely available software. I used a WordPress plugin, but it didn’t work quite the way I wanted it to, and didn’t look at all the way I wanted it to. I worked at it until I had something that was good enough to get started.

PayPal is not hard to use, but it is not very well documented. And the ecommerce software isn’t documented at all. I bought my own product about four or five times to verify that the system works, and works as well as possible, given the limitations of all the various components. I set my product price very low to keep transaction costs to a minimum, since transaction costs are what it costs me to buy my own product.

If you are considering the idea of getting in to ecommerce at some point, get a PayPal account right now . You need to have one for at least 90 days before you can make full use of all of its features.

For marketing , I am using Google AdWords exclusively. For now. Although setting up and using an account is straight forward, AdWords has some very strange characteristics that I am just learning about. For example, I chose a key word that worked pretty well, but then Google, in it’s infinite wisdom, or lack there of in some cases, upped the bid price on my key words, citing quality issues, in other words the Google system determined that the best key phrase did not very well match the landing page or the ad.

For about 6 to eight hours, I was getting my ads served and very inexpensive clicks, but as I say, Google decided to change the price on me. Things change that fast. One phrase went up by five times, then later up by ten times. Another went up by 50 times. I’m not planning to bid that high at present, I’d rather see what it takes to get the bid price down again. As much as Google changes the rules hour-by-hour, I may change my mind as to what I bid, what ad network I use, and what other promotion techniques I use hour-by-hour.

This is all very interesting. And eventually profitable, that is why I am doing this. The question is: how to make it profitable?

I got a few clicks yesterday, and a few people downloaded the sample chapter. The next challenge is for visitors to buy the book! There are endless adjustments I could make to the website to get it to convert better, including reducing the number of steps to go from landing page, to shopping cart , to checkout.

Each of the above paragraphs can be expanded into many many posts. I’ll certainly be sharing some of what I learn as I go along.

Note that since there is so much I could write about, but won’t be spending all of my time writing this blog, your comments and questions will help direct what I write about next.

I recently ran across another article about whether to use rel=nofollow in links from your visitors comments. Jonathan, the author, advocates installing the Dofollow Plugin For WordPress, which removes the rel-nofollow attribute in comments.

Is this a good idea?

In a previous article on this blog, I commented that using rel-nofollow may improve the quality of comments. While this may be true, it may also discourage commenting in general, certainly among the link-savvy. At the same time, it may discourage comments for the sake of a link to the commenter’s blog.

I think I have a solution to this for anyone who wants to develop a plugin to make the following idea happen.

Extending the publish/delete comment function in a blog, how about being able to publish with or without rel=nofollow? This way you can reward on-topic comments with links to a related blog, but still publish comments that are a bit off-topic, or do not link to a related blog. Such a plugin would flag which comments are which, and publish your rel=nofollow policy next to the comment form.

I have not yet installed the dofollow plugin, but I am considering it.

What is your opinion about rel=nofollow? Would you like to be able to rate the comments left on your blog and reward the most relevant ones? Are you more likely to leave quality comments on a blog which does follow links back to your blog? Your comments please!

More articles about rel=nofollow:

Burry rel-nofollow - clearly against it.

I started looking into ecommerce solutions a while back, and wrote about my initial research in an earlier article about ecommerce systems for digital downloads, in which I discussed a few systems I have run across.

So far, I have not found the ideal system. All of them have something that is not quite right, often very not right.

Zen cart, OScommerce and the like could work pretty well, but they are designed for a complicated store with physical products, and this sort of system is not really ideal for digital downloads in my opinion. I felt that it would take a tremendous amount of work to learn the system well enough to get the template and buying process to work the way I want it to. I already know a fair bit about WordPress and Drupal, and want to build on that.

Drupal has two options, and a third on the way. Neither e-commerce, nor Quickfile really have a properly thought-through buying process that is ideal for the buyer. E-commerce really would be good with some streamlining, and probably I will roll up my sleeves and do just that eventually. Mean while, it has a very long list of unresolved bugs which makes me leery about investing time and energy into adopting and deploying it. Ubercart looks very promising. Handling digital downloads is on their to-do list, but it does not sound like a priority for the development team, unfortunately. I suspect it will be pretty good though, when it does become available.

The best (least worst) I have found so far is wp shopping cart for WordPress. It has some problems too, as I will explain, but it 1) works and 2) is possible to provide a pretty good buying process for visitors (with a fair bit of work). It did not work the last time I tried it, but they have fixed a few things since then.

My plan is to start with wp shopping cart for WordPress , then determine if my digital download store concept works, or more to the point, how it works and how to optimize everything from the product selection to marketing to make the on line store really successful. At some point, I will deploy a new platform, probably Drupal-based. At that point it will be worth putting in considerable effort to rework the software to provide an excellent buying experience to visitors.

Customizing WP Shopping Cart

WP Shopping Cart is functional, but fairly ugly out of the box. The trick is to do extensive work on the CSS for both the WordPress theme, and for the shopping cart system. For example, rather than just having links for various functions such as visiting the store or proceeding to the check-out, why not make them look more prominent and more like buttons? I gave various buttons a height, width, background colour and border to suit the prominence I wished to bestow upon them.

Some elements are very hard to specify with CSS, necessitating some code hacks to add classes to some elements. I was able to circumvent much of this by applying additional class tags within the language files.

In the end, I got a system that looks pretty good. (It’s not completely ready, so I am not linking to it from here, not yet anyhow). I have discovered some major technical stumbling blocks with WP Shopping Cart however.

The biggest problem is that the system does not fully automate the buying process. Even though it is a digital download, it will still require my intervention in some cases. When this becomes too time consuming and troublesome, I will certainly switch to another system, but it will be good enough to get started.

Another shortcoming is that if the user has any trouble with the download, there is no way to give them access to the download page.

I have discovered that by looking up the session id in the database, I can provide a link to the buyer’s download page in a very manual, back-end sort of way. If I get many questions about this, I will provide further details on how to go about this.

They system I am setting up is close to ready. More posts to come on this subject.