Underutilization of Databases

[billbenson]

I started responding to another thread, but thought this deserved a thread of its own.

In site design threads I have seen here and elsewhere, there is a concentration on the importance of layout, graphics, color selection, SEO etc. I don't recall anybody ever mentioning the power of databases and other server side scripting.

One of the problems is it is very hard to present this in a manner that someone in need of a web site can understand. A simple example is a banner ad that says "find girls in Dallas" when a Dallas user accesses the site.

There is talk here of good targeted copy, landing pages (by various definitions), and countless other things. How about targeting that copy at that particular visitor.

You can save things to a db like the search term in google that brought him to you, and countless other things. From this, you could make a good assumption that it is the same visitor (ya, dynamic IPs change, but usually on the D block and combine that with other factors...). You know he was searching for product X the last time he was there. This time you put more information in the side bar on product X, a free shipping offer today only, or whatever other things you feel that will help close that customer. There's all kinds of stuff you can put on the page to help qualify him and never ask him to fill out a form.

What's this allow you to do? It allows you to present a landing page, not just for a product, but for a customer. It also allows you to profile customers to figure out price points, purchase urgency, and a mountain of other things.

For those interested in a web site or improving their existing site, I would suggest doing some research on what can be done to capture and save data and how databases work. You don't need to program them, just know what they can do. You may find that a well used database can dramatically improve site performance.

[vangogh]

Presenting different content to different users has been gaining more traction in recent years. Within the WordPress community there are now a variety of plugins that alter content based on the visitor.

For example many people use a plugin called greetbox (http://wordpress.org/extend/plugins/wp-greet-box/), which lets you welcome people based on where they came from and if new to your site suggest they subscribe via rss. The plugin can also detect keywords from a search engine and present related posts based on those keywords.

Other ideas I've seen talked about are to not display ads when people visit from a social media site as those visitors tend to be turned off by ads more than others.

I think part of the reason you don't see much talk here of databases and server side scripting is because few of us could even get into the conversation. I don't know how many here currently work with databases and server side languages, but I don't think it's a lot. Personally I'd welcome these discussions, but I'm not sure how well they'd do in general.

I think for the majority the use of a database is simply having a content management system powering their site. Most people aren't concerned with the details of how their site works as long as it works. Still you make a good point about how a site's backend can working in tangent with the site's marketing. Even if most don't want or ever need to understand the details of databases and server side scripting, it's probably a good idea for them to understand how both can help with building a better site.

[billbenson]

Even if most don't want or ever need to understand the details of databases and server side scripting, it's probably a good idea for them to understand how both can help with building a better site.

And that is really the point I am trying to make. The business manager doesn't need to know how to do it, but if he does have a good idea of what can be done and provides you with a better spec, you can build him a more successful website.

As a designer, if you could put together info in laymans terms on what can be done server side to improve the performance of a website, it may be a good pitch, although I suspect most clients will have unrealistic expectations on capabilities and prices as always.

[thx4yrtym]

Bill / vangogh ,

Perhaps if others understood some of the possibilities there would be more interest.

If I'm not creating custom pc-based programs for growing businesses, I'm creating web-based applications that are usually an order entry system where the client needs their customers and or their reps to have access to the system from anywhere. Currently working on a web site in the education field that will be subscription based. When the client was given a choice of web or pc-based and I explained the advantages of each, they chose the web system.

I guess folks need to understand that it's possible to create nearly the same functionality today on a website as any stand-alone pc-based program.

The fact that there are so many open- source tools available to us developers means that the cost to the client isn't nearly as prohibitive as it was years ago.

I totally agree that databases used for the front-end of a web site add all kinds of possibilities from a marketing stand point, however, that's not the end of the story.

[billbenson]

And for those that are doing their own sites and may have shyed away from programming:

I conned my way into my first web design job after spending two weeks with a DreamWeaver for Dummies book. Within a couple of months I could see the power of databases and bought a php / mysql for dummies book. I could write php in code, long before I could write html in code. Cart before the horse,

For those doing sites for themselves, I strongly recommend digging into it a bit. You might be surprised what you can do.

[vangogh]

Perhaps if others understood some of the possibilities there would be more interest.

I think most everything is possible. It usually comes down to a question of how much you're willing to spend to make it happen. It's true that open source solutions greatly reduce the cost, there's still a cost. It's hard to tell people in advance everything that might be possible, because the list would be endless. Still we can tell people some of the things that are possible to get them thinking.

Bill I agree if you're designing or building sites on your own you should start digging into the html, css, server side programming, client side programming, and databases. You won't pick up all of it overnight, but you can pick it up. And if you don't want to dedicate yourself to learning all of the above it's still a good idea to become familiar with each and develop a working relationship with someone who is an expert.

[painperdu]

You make good points but you make it sound easier than it really is. It's hard enough guessing what someone wants in person, anticipating what they may want by data-basing what they did in the past is even more difficult.

Google and the bigger sites are starting to get into behavioral-based, context ad serving. I think we'll have to watch them and learn how they approach this. They have a large enough user base to do it.

[vangogh]

Some things you're not going to know and so the best you could do is guess, but there are things you can know about any visitor and adjust. One is their relative location. Server side languages give you access to the IP address of the visitor so you could tailor things somewhat based on location.

I mentioned greetbox above. It lets you adjust based on the site the visitor came from. Sometimes that won't mean anything, but in some cases it will. As I mentioned social media users tend not to be ad clickers so if you serve ads, maybe when you detect someone visited from a social site you can use the space for something other than ads.

You can also capture the keywords a visitor used if they're visiting from a search engine. You could then change your page heading based on their search or perhaps highlight the keywords they used on your page.

You're not going to be able to predict everything they want based on a few things, but there are ways you can tailor your site based on information you can determine about your visitors.

[billbenson]

Smaller sites can also do some things Google can't. I can, for example assume that an IP with a different D block is probably the same customer with a dynamic IP because the raw number of visitors is low. Google couldn't do that.

Amazon shows results based on your prior visit, but they have your IP or a cookie set.

And you are also correct, it's not easy. But there is a lot of talk here about how to improve closure rates, and everyone I recall has been revolving around static pages. People should be aware that there are some very cleaver things you can do with dynamic pages.