And it does make a good story, with lots of false starts and lessons learned—as well as obstacles overcome. And there is stuff here to interest anyone who has contemplated moving or reconstructing a Web site. What is more, if anyone is in the same predicament, looking at a Web site based on a proprietary CMS, I have a lot to pass on.

But now I am in another predicament. I am too busy to go into the detail. To give you an idea: I have been building a new Web site for a local organisation which already has a Web site. We agreed a date in mid-March for the switch. But the organisation’s old Web site has vanished from the Web! They are now asking me if they can switch this week.

So: here is a summary of what I did to reconstruct the proprietary CMS site.

1. I created a MySQL database, together with 4 tables
2. I populated the tables partly with data that the previous designer kindly handed on to me, in the form of an Excel spreadsheet, but mostly with data that I copied and pasted from the site.
3. I created the main ‘site’ pages as static pages, simply copying the HTML that the previous dynamic system generated.
4. I created the ‘browsing pages’ with a set of HTML templates and PHP scripts.
5. I downloaded, one by one, by hand, around 800 images, including property images and icons, and put them in appropriate directories.
6. I created a PHP script that displayed any of the 70 or so properties, taking descriptions and prices from the database, and images from directories.
7. I created a wealth of interactive features: a search facility, a registration form, popup enquiry and print forms, a ‘lightshow’ image gallery, an availability calendar (for each property), and various booking forms.

At last a Web site was there, which looked very like the old one. At this point I turned to creating a new client interface, so that my client had a new personalised CMS.

As you may guess, this turned out taking as long as reconstructing the Web site visible to the public. But I was able to give my client a much more friendly and intuitive interface, with many more facilities than she had had before. (Oh, and as part of the process, I had to recreate the database tables, to make her job easier…)

I would point you to the result, but the public part of the site is only my reconstruction of the original designs, and still contains the mangled CSS and HTML generated by the old CMS.

And nobody but me and my client can see my lovely new client interface. Sad.

There is a general summary of the position on an Australian Web site. This is (part of) what they say about CMS sites:

“Content Management Systems – a mixed blessing
An alternative to a personal web designer is a CMS or Content Management System. The main benefit of a CMS is the ability to update your site yourself – within certain parameters – over the internet, and rolling all your costs into one monthly payment. But you need to go into a CMS contract with your eyes open.
1. CMS do not provide distinctive sites. They are, almost by definition, formulaic. Look at some and see what you think.
2. You don’t own your CMS site in the same way that you own a site designed by a web designer. With CMS you own the content but you cannot walk away from your CMS contract with your site intact – your site can only operate on the CMS operator’s platform.
3. With a CMS website you are locked in – you cannot change to a hosting company that provides more competitive rates or better features without losing your website.”

Point 3 in effect repeats what the English company say about their own sites.

So what was I to do when I was asked to ‘move’ exactly such a Web site, by a potential client to whom I had been recommended—by people I would prefer not to let down?

Of course I said ‘yes’. I committed myself to recreating the site, i.e., constructing a completely new site that would look just like the client’s current site, with a completely new database, a new set of scripts (PHP rather than the current ASP), all the existing images (several hundred of them) and a client interface that let her do everything she had done with the old CMS.

How did I get on in this foolhardy enterprise? Answer in Part 2.

Belated, because I have been so busy that I have neglected this blog. However, I now have a couple of posts on hand which I should complete in the next week: one on moving a Web site based on a proprietary CMS (that’s been my most time-consuming job) and one inviting comments on (another) new Web site.

Watch this space!

I have a couple of Web sites in this position right now. One is for two karaoke presenters: brookies karaoke. The other is for a friend who wants to sell her unique spacious villa in Calpe. I am doing plenty of work behind the scenes, which the website owners can check out, but neither of the full Web sites are ready yet to go public.

Obviously, you can’t go anywhere further on the site from the home page of a one-page Web site—but at least you can find contact details. What is more, from a website designer’s/owner’s point of view, the minimal Web site is available to be noticed. (And, to my astonishment, Web search engines often find one-page Web sites quite fast.)

1. The Web site has got to be found.
2. The Web site has got to be attractive.
3. The Web site has got to deliver.

OK. You look at these wishes, and you think, “How obvious! This is just common sense.”

So how come so many Web sites fail on at least one count—and some on all three? Let’s look at what may not be so obvious.

In the first place, the Web site has got to be found by the people you want to find it. That means designing the whole Web site for those specific people, and calling on SEO and marketing skills to catch their own specific keywords—every time you add text.

In the second place, the Web site has got to be attractive to the same specific people—not to the website owner, not to the website designer, but to the website visitor. You have to deploy the colours and the layout and the design which will engage the visitors you want, and make them want to explore your Web site. If we were all attracted in the same way, we would all be pining for Marilyn Monroe. Personally, I was always more attracted to Jane Russell. (You get the point.)

Finally, the Web site has got to give the people that you want to visit exactly what they are looking for. They came to your Web site because they were looking for something. You can bet every cent you have that they weren’t looking for an advertisement. Advertisements are what people fast forward through when they record TV programmes. So why would they be looking for one on the Web? Just give them what they want—no less, and (this is really important) no more.

So there you have it, and now you know what I mean. Your Web site has got to be found. Your Web site has got to be attractive. Your Web site has got to deliver.

Is Web hosting an area where it is right to assume that money and quality of service are connected? Let’s look at the facts.

There is no real mystery about Web hosting. You rent space on a computer which is permanently connected to the Internet. You also get a bandwidth (traffic) allowance. In general, the bigger the space you rent, and the more visitors to your Web site you get, the more you would expect to pay. Just as if you were renting space in an office building.

This is the case with my Web hosting plans on Web Costa Blanca. And my prices are very low, even for the biggest allocations of space and traffic. So what else are people paying for?

Two things: (1) a range of extras; (2) service.

1. I offer an extensive range of extras (personal control panel, email accounts, programming languages, site builders, databases, etc., etc.) This blog is not the place to trumpet them all, but anyone who cares can find them in the Web hosting plans section of Web Costa Blanca. I have looked at dozens of other Web hosting sites, and never seen a range of options to beat them.
2. I offer a one-to-one personal service to all my clients, who can reach me by email or telephone whenever they hit a problem or want to ask a question.

Why do I charge so little, then? It’s probably something to do with prejudices left over from my radical youth. I’m happy to charge for my labour—website design, website development, website maintenance. However, even when my clients have access to so many tools and extras, I’m not happy to make a profit on renting space.

And as for the service, how else do you get job satisfaction?

Let me admit, to begin with, that many website templates are elegant and professional.

I still don’t like them. And here are 5 reasons why.

1. Elements of the template—and sometimes the entire template—may be inappropriate. The imagery may not be right for this particular Web site, or the colour scheme, or the layouts, or the navigation. In extreme cases an entire site has to be created following a formula dictated by a given navigation scheme. Or the number of top-level pages is limited by a given set of navigation buttons. Visitors may find it harder to do what they want on the site, because the template isn’t designed for specific visitor tasks.
2. Even where the look and feel of a template is OK for a particular Web site, and visitor tasks are not hampered, the website template may not express what is most important about a business or organisation—its own character, its unique selling point.
3. There is no real design input from the client—no chance for her or him to stamp their own wishes or personality on their own Web site. The best they have is a choice between templates.
4. There is no real call on the website designer’s craft. The client gets a supermarket ready meal instead of a menu of chef-prepared dishes.
5. Most important: the template Web site will not work for the client as it would if it were the result of serious preliminary analysis. Before design comes analysis: what will count as success for this particular Web site? what visitors does the client expect or want to attract? how are these specific visitors most likely to join in the campaign/sign up/part with their money (whatever)?

No template-based Web site will ever work for a client the way a bespoke Web site can.

I recently signed up for a local business directory. Since I am using it as an example of what not to do, and since the people who run it are such nice people, I won’t name it. When I looked at my entry I saw the word Website, but no website address. Also the word Email, but no email address. I wrote to the directory to protest.

Before they wrote back (actually, they haven’t written back yet) I checked my entry again. Then I thought, ‘Maybe the bare names are links.’ And so they are. By clicking on the word Website you can go to the Web site. Click on the word Email and you can compose an email.

So I was being stupid, right? You can make up your own mind on that. The important point is that it is the job of the communicator to communicate. If a Web site doesn’t communicate, to every relevant visitor, it fails. Here, the bare words were not enough to communicate that there was hidden information. There was no underline to indicate a link, and in 2009 no colour is enough. (Check out my earlier posts about links.) As it turns out, there’s even a tool tip—but you’d have to move your mouse to the words to find that out. So this is rule 8.

8. Be 100% obvious.

And 100% means 100%. Everything a website visitor can do must be crystal clear. All information must be up front and visible. There are plenty of ways of making even an email address visible to human visitors without its being visible to spambots. And if you hide it, you must say so, in an obvious place that a human scanning the page will see at a glance.

If a website visitor has to guess or work out what to do, the Web site is not 100% obvious. If a website visitor has to do something off their own bat to find information that is on the page, the Web site is not 100% obvious. Remember, 100% is 100%: 99% won’t do.

The site was lovely to look at, with elegant illustrations, sophisticated typography, and plenty of white space.

Of course, even as a visual design, it had a couple of problems. The menu system wasn’t exactly proportioned, and the balance of the page wasn’t quite right. But that wasn’t its real problem.

The designer had concentrated on making visually attractive pages. He hadn’t spent time thinking about the purpose of the site, who he really wanted to visit, what those visitors would be looking for, etc., etc. There was no focus, no site concept other than to look good, and it was barely useable.

I have written before about how website design is not about creating pretty pages. In a follow-up post, I compared website design to designing teacups or bridges, but in formulating my response to this designer’s site, I was hit by a much better analogue: kitchen design.

What really counts about a good kitchen design is that everything a kitchen needs should be in the kitchen, that the person who is going to be using the kitchen can find everything easily, that everything is really easy—even a joy—to use. And what matters about the look of the kitchen is that the kitchen user, plus her/his friends, feel at home in it, enjoy being there. (The designer is not going to be cooking in that kitchen, nor drinking coffee with friends, nor having breakfast, etc.)

You can spell out the similarities for yourself. Once again it underlines how good Web design is visitor-centred Web design, not client-centred Web design, and certainly not designer-centred Web design.

I have recently finished the Web site: for an Indian/Nepali restaurant in Calpe. You can now inspect this Costa Blanca restaurant Web site yourself.

A fellow Web designer that I met in a bar in Calpe suggested I should use a template for such a site. It would have been easier, and maybe quicker, but I wanted to give the owners an individual site with a character of its own. You can once again check out my design notes for it if you browse my website portfolio, on my personal Web site.