From yesterday’s post:

The methods that I know of are as follows:

• Search Engine Optimization (SEO). Customers come to the site because they have posed a question to a search engine, and one of the results has directed them to the site.

Briefly looking at the first method of marketing and publicizing a web site, it must first be clearly pointed out that Search Engine Optimization (SEO) is a huge science in itself. Thousands of books have been written on this subject by people far more knowledgeable than I. All I will attempt to do here is to put down some of the most important lessons that I have learnt. My knowledge on this subject is very limited and if I make any mistakes here, please feel free to correct me via comments.

The first lesson is that the rules for search engines are continually changing and the search engine corporations do not publish a definitive guide on exactly how the search engine evaluates your site. In a sense, one is therefore shooting at a moving target. It also seems clear that the guidelines laid down in the extensive tools and documentation provided by Google-for-Webmasters do not necessarily apply to the way MSN or Yahoo evaluates and ranks the importance of your site.

Each Search Engine corporation offers hints, tips and guidelines about how to make your site “Search Engine Friendly”. However, for some reason which eludes me, they seem to hide this information making it sometimes quite difficult to find. Here are a few starting points:

For Google, go to:
http://www.google.com/webmasters

For MSN, go to:
http://webmaster.live.com/

For Yahoo, go to:
http://help.yahoo.com/l/us/yahoo/search/webmaster/webmaster-01.html

And for DMOZ, the open directory project, you’ll have to hunt around because the help for webmasters seems to be very scattered. As a start, go to:
http://www.dmoz.org/help/

The next thing to take notice of is the concept of Google’s sandbox. It’s an undocumented aging delay that Google seems to penalize some sites with. You should read up on this at: http://www.rightclickwebs.com/seo/google-aging.php. Moral of this story:  - submit your site to Google ASAP and continually resubmit as its content gets built.

More about SEO on Monday…

We have a functional site in place. Now we need customers.

• The virtual electronic doors are open and the site looks beautiful.
• There are courses available for sale (not a large number at the moment, but growing every day).
• The banking infrastructure is in place - we can accept credit card payments.
• The Tusker shopping cart is tested and functional.
• The course shopping cart is tested and functional.
• The customer can find the courses using various different methods.
• Each course has its corresponding description and sample pages.
• The small print and legal stuff is in place.
• The log-in and password system works.
• In short, the virtual shop is functional.

But where are the customers? They are out there, for sure, but they just don’t know about Fundisi.com yet. It’s no use having this wonderful shop if no customers are, virtually, running in and out.

In modern-day commerce, marketing often starts before the product even exists. Companies try to get the edge on their competitors by selling  “vapourware”.  I’ve always found that idea intriguing, but somehow not right. I wouldn’t go so far as to say ‘dishonest’, but I feel you must know that you can deliver what you say you can and have confidence in your product. How do you do that if the product doesn’t exist yet and the vehicle to sell it (the virtual shop) isn’t functional?

Well, now that everything is in place, at last, it is time to start the marketing phase.

How do you get your product to the Internet marketplace? How do you get customers?

The answer is easy, but the implementation is not - you need lots of Internet traffic. People must come to the site in their droves. If the product is good, they will buy, but no matter how good the product is, if there are no customers coming to the site, no one will buy.

There are a number of different methods to bring traffic to the site, and there are probably other methods which I have not thought of. (Note, Dear Reader, if you know of an effective way to get customers to the site which is not mentioned here, please, immediately, reply to this post by leaving a comment.)

The methods that I know of are as follows:

• Search Engine Optimization (SEO). Customers come to the site because they have posed a question to a search engine, and one of the results has directed them to the site.
• Newsletter and Email list building. Gather the email addresses of potentially interested customers and send out regular newsletters enticing them to the site.
• Viral Marketing. Physically speak to people about the site, or use social networking sites like Facebook. Entice people to tell their friends about the site.
• Pay-per-Click advertising. An advert is placed by a search engine company and whenever the link to your site is clicked you get charged. No click, no charge.
• Banner/display advertising. A more traditional form of advertising where you pay for space on a busy site and hope that your ad will be clicked.
• Get other sites to link to yours. This is like a mini-ad on someone’s site or an entry on their blogroll, but usually necessitates a reciprocal link. (With our heavy graphic design, this is a bit difficult to implement on an ad-hoc basis.)
• News/Press releases. Trade related sites are always happy to tell their readers about new products.
• Blogs that offer some useful content to attract visitors. You can then put as much advertising as you like on your own blog. Like this one for example, or my other photography blog.

Over the next few weeks, I will blog about these options in detail…

While the advantages of a beautiful design with heavy graphics are obvious, this wonderful visual effect comes at a price. Speed. Heavy graphics make each page load distressingly slowly. The more images on the page, the more bits of image data have to come down the pipe to your computer before the image can display properly.

The Complexities of Slow Browser Speed made Simple

It works like this. The contents of a web page fall into one of three categories: Text that you see reproduced on the page; text that you don’t see; and graphic images.

1. Text that you see provides the words for the page and is usually selectable and copyable (but not modifiable) and is similar to the text in a word processor.
2. The text that you don’t see provides programming instructions to the browser and metadata that search engines can read (like the page title, description, keywords and technical messages to the search engines).
3. References to Graphic Images must also come down the electronic pipe from the server far away to your computer at least once.

All three types of information merely consist of a very large number of “packets” which contain the electronic binary information. (It gets very technical here, but we only need an overview to understand what is going on.) The main thing to grasp is that text (in its visible or invisible form) doesn’t take long to download while graphic images take much longer.

The reason for this is that the text which downloads merely instructs your browser to use a text font which is already installed on your computer to display a character. The font itself may be quite large, but it is already located on your computer (put there when your operating system was originally installed) so it doesn’t need to be downloaded again. Text is therefore quick to download, and web pages with lots of text and very few graphics load very fast indeed.

Graphic images, on the other hand, first have to be downloaded to the computer before they can be displayed, and because they are quite large, they take a while to come down. So, heavily graphicked pages are consequently slower to download, and can make the page so slow that the user gets irritated and clicks away.

So, a compromise must be struck between lots of graphics and page loading time. But what if you just have to have a heavy pictorial presence like we have on Fundisi.com?

Solutions

There are three tricks that we used to reduce download time:

1. Graphic images can be optimized in Photoshop to display fewer colours by tricking your eyes. This is a whole science in itself.
2. The design can be broken up into a series of “image tiles” which gradually display as soon as they have been downloaded. This results in an effect where the image is gradually being built up, and the user realizes that the site hasn’t frozen and is being “entertained” as the image is being assembled. This effect and the actual order in which the image tiles are displayed can, to a certain extent be controlled by the programming of the web page.
3. Large areas of uniform graphics can be split into small identical pieces. If these pieces are absolutely identical, they need only be downloaded once and the browser just references the same image again and again. This means that download time is saved and a large area can be “painted” quickly by the browser.

All this web page optimizing of course takes time and effort - which increases the cost of the site. There is, unfortunately, another not so obvious consequence to this: A site page with a large amount of graphics is not easily modified. The design must be right before the page is built. This can, to a certain extent, be done by approving the original Photoshop design, but then little things like input boxes and drop-down lists which are not graphic elements behave differently in different browsers.

Text in different browsers takes up an unpredictable space (especially in the new browser version that is released the day after the web page is created!)

If all the text were rasterized into pixelled images, this would cure many such problems, but then the page would get slower still. It’s all a process of compromise and frustrated tweaking. It’s no wonder web site programmers tend to have frayed nerves.

The Zen-Cart “Default” Product Pages vs. Fundisi.com’s Product Pages.

The original Zen-Cart product display system is relatively simple and the style is instantly recognizable to someone who is familiar with the on-line stores created by the Zen-Cart software. A typical “Zen-Cart look” is a 3 column home page, with Categories in the left column, information in the right-hand column and “featured products” in the main center column as shown here:

A Typical Zen-Cart-Style Page. (Click here to have a look at the original page)

Briefly, the “populating” of the store stock works like this:

1. The Administrator accesses the back-end of the site through a special web interface.
2. A new product is added by filling in a form with the relevant data - description, price, metadata etc. A picture can be uploaded. This information is automatically added as a new record to the database, and the product is immediately available for sale. This is the beauty of a CMS (Content management System).
3. The software then allows users to use a virtual shopping cart to select the products from a hierarchical menu system and pay for them via credit card.

Because we wanted a very different unique “look and feel”, but still use the CMS system, the original software had to be modified to an outrageously large extent. (Zen-Cart is open-source software so one is legally allowed to do this.) Our trusty programmer put in countless hours pushing the Zen-Cart code to its limits. We wanted a reference menu on the extreme left, a scrollable course listing on the centre-left of the page and a course description with a couple of example images on the right hand side:

Our Main Product Page which Resembles an Open Student Folder.

If you look at our actual course page in more detail, and then compare it to some of the showcase Zen-Cart designs, you will see an enormous difference in how the products (in our case - downloadable PDF courses) are displayed.

Unfortunately, all this graphical “eye candy” comes with a speed price…

Last week’s blog post ended with:

However, unbeknown to us, the graphic designer would get her way after all, in a turn of events that none of us expected…

If you’ve been following this blog from the beginning, you may recall the blog post I made about the Search Engine Friendliness of the URLs, and our subsequent solution to that problem. This solution also settled the argument once and for all about search-box vs. menu system, because SEO is very high priority for us.

There are two kinds of menu systems:

1. The first is an array of high-level categories, each category unfolding to more specific lower-level categories. Sometimes these menus are fully visible in their hierarchical tree structure, but above a certain complexity, they would need to be hidden, and expand with a deliberate click or “mouseover” movement. These menus are either built using CSS or can be built using a DHTML menu creation tool which writes all the javascript for you. An excellent example of such a menu system is called Allwebmenus and is available from a company called Likno Systems, but we didn’t go this route. Instead we went with the second option below.
2. The other sort of menu system is where the actual web pages themselves form a hidden hierarchical system. The first (top) page contains a list of the major categories, and when one of these categories is clicked, the appropriate second-level page appears, which in turn has further more specific clickable links.

Since for search engine optimization reasons we had to have a series of descriptive web pages, with the lowest level containing the actual course details, the menu system in option 2 above was obviously the way to go. So our Graphic Designer got her wish and actually got two menu systems. The first being our three tiered system of:

• Home page with main categories.
• Second level Category pages listing the courses available within that category.
• Third-level pages each containing a specific course’s details and example picture, designed to be indexed by a search-engine.

The second one was the alphabetical course listing built into the Zen-Cart section of the site. This course listing is automatically re-built when a new course is added to the database. More details about this process tomorrow…

How is a customer supposed to find a course? One approach would be to type in a word or a phrase and the site engine would display a list of courses matching the supplied phrase. Another more traditional approach would be to have a hierarchical menu system where you “drill down” through categories and sub-categories until you come to a list of available courses within that section.

If you know what you are looking for, using a search box is far more efficient than menu browsing. But what happens if you don’t know what you are looking for, or if you just want to browse what is on offer? Then a search box might not be the best solution.

The team argued long and hard on this one - each passionately putting his/her point of view.

The graphic artist wanted a traditional menu system something like this:

Menus on the Right of a Course Description List

I argued against this on the grounds that when we would eventually have so many different categories, where would they go? I wanted something much simpler - just a plain search box like the simplicity of the Google search page:

A Simple Search Box

I was howled down with scorn. We argued and fought and eventually reached a compromise where there was a search box and a sortable list of courses which could be browsed. Take a look here at what we finally settled on.

A Note on Management Style

I have always felt that while the final decision rests on my shoulders, if I had gathered together a team correctly, the ideas and opinions from a competent creative bunch of individuals could well be better than my own, and that it would be foolish to stifle such a debate. At times, passionate argument is more productive than autocracy.

However, unbeknown to us, the graphic designer would get her way after all, in a turn of events that none of us expected…

The traditional concept of a web site defines the home page as the main “front door” to the site. Nowadays this is only partially true - via search engine links, the site can be entered via any page that appears in a search engine’s results page.

So, if a customer happens to know the site URL and types it in like this:  www.fundisi.com, then he/she will enter the site via the home page (sometimes called the index page). If the customer discovers the site via search engine results, the chances are much higher that a page containing course details (a second- or third-tier page) will be the first page that appears in the browser window.

So, how much importance should the home page actually have? As explained above, it depends on where you think your customers will come from. What about customers that respond to an e-mail sales campaign or newsletter?

Do you direct them to the home page which then acts as a springboard hub allowing the customer to manually navigate to the page he/she is looking for? This has the advantage that the customer might become aware of other news items and new courses recently put up.

Or does your sales-letter link the customer directly to the relevant page? That is the most efficient method for a customer who is only interested in the subject that is clicked. There is an old saying that it should not take a site visitor more than four clicks to get to the page he/she is looking for.

Is this still valid? I think not…  Customers want less than 4 clicks - they want to get directly to their target with just one click.

And when they want to just browse? What’s usually the first page they go to? You guessed it - the home page!

So, I conclude that the home page still has its place, but each page that is indexed by the search engine should nowadays also be able to do double duty - as a detail information page and as an informal “home” page with a clear navigation link to the “main home page” - the front door to the site.

The way that Fundisi handles money is similar to that of the casino. In the casino, you first go to a secure area and purchase your chips with real money. Then you move into the casino with “money” that is valid within the casino, but isn’t real outside the jurisdiction of the Casino. There are two reasons for this:

• Psychologically, you aren’t playing with real money so you tend to gamble away your money more easily.
• You have physical uniform chips that the different machines can recognize and operate with.

With Fundisi, the result is the same, but the reasons behind it are different. You first need to buy Tuskers with a credit card. Your currency may be different to the previous customer, so how on earth would we list price the courses? In 99 different currencies? Not very practical. So we needed a single currency which wouldn’t disadvantage or offend anyone. Our site currency is therefore called the Tusker and is exactly equal to 8 South African Rands, or, at today’s exchange rate, is approximately equal to ¾ of a US dollar or nearly ½ a British Pound or roughly …

Each Tusker buys one course. Your credit card banker does the conversion from South African Rands invisibly and all very accurately, so you never get ripped off. All our “real money” shopping cart pages are distinctively violet and yellow in colour and are of course located on the secure part of the site.

The second reason why we need to have two shopping carts is because the courses are so inexpensive. The bank-charges and commissions become proportionately too high when you are spending less than one US$ per transaction. So we have to make the minimum number of Tuskers at least three in order to be economically viable.

The third reason for this dual shopping cart system is so that we can offer discounts and specials for customers who wish to buy many courses.

So, having bought your Tuskers, we need a different Course Shopping Cart to enable you to buy the courses with Tuskers. This also takes place in the secure part of the site, but this time the pages are in the usual grey/orange/white combination that runs through the theme of the whole site. The unspent Tuskers are kept in the customer’s account for future use. The Fundisi Elephant won’t forget about them!

So, it’s a little more complicated than at first glance!

One needs to be very careful about the emotions that a particular image might stir up. Take, for example, the case of lips and the effect on a log-in screen.

Initially, the design of the log-in screen was this:

The initial log-in entrance page graphic

One of the team members had this comment:

I’m putting the cat amongst the pigeons here!!

Underneath the ‘Bon Appetit’ is the picture? of the blotted lips. We need more of a smile of appreciation and satisfaction to come across.

To my way of thinking there is no emotion transmitted by this picture and a valuable opportunity may be wasted.

Pity we can’t add a burp.

To which the graphic artist responded:

What an excellent idea!! <rummaging for lipstick> I shall amend. <looking for lippy smiley, but with no success.>

She then posted the following suggestion:

The modified Log-in page

She requested comments on the artistic merits. To which I responded:

You’re asking a bunch of non artists … hoo ha…

My feeling: those lips are a bit smug - like I’m going to sucker you.

I dunno…

I think you’ve got a smirk there.

The graphic artist then posted a whole bunch of lips images for the team to choose from:

A selection of lips

After much forum discussion of the psychological effect of the lips on the napkin, and whether the lips should be open or closed and by how much, the graphic artist posted the result of the “design by committee”:

The final friendly, honest and open lips

Since that discussion there have been other design changes to the actual, log-in page, but the lips have remained the same. To see the final result as it appears today, click the actual log-in page on the site and see what you think.

When we happily speak of an Educational Course at Fundisi.com, what do we actually mean by that?

The name is something of a misnomer since it suggests travelling down a step-by-step path of education and heavy study - conjuring up visions of piles of text books and fat instruction manuals with an exam at the end of it.

This mental image is a far cry from what we are actually producing. Our basic course concept consists of:

• A short (and hopefully sweet) article on a particular subject - between 10 and 20 pages long.
• An easily downloadable file that is usually not printed out (although this is perfectly possible if you hate trees and want to do your bit towards using them up) but rather retained on disk.
• Information which is supplied in a palatable form which only requires half an hour or so to read.
• Inexpensive Education.
• Authoritative information written by an expert.

The whole idea of Fundisi.com bases itself on the premise that we all need information, but we just don’t have the time to read a reference or text book from cover-to-cover when we are only looking for a particular segment of information. However, that information must be correct, reliable and authoritative.

To take a look at an example of such a course, click here to directly download a sample.

Gradually as time goes by, there will be more and more courses available on the site, and within a few years we anticipate having a huge repository of information. Since everything is digital in format, courses can easily be kept up to date. Another advantage is that when a customer has purchased a course, it (and all subsequent updated versions) can be downloaded at any time as often as is necessary.

Now that’s what we call a “Super Course”!