GOODIES TO GO! (tm)
June 21, 1999 -- Newsletter #33
Full Text Search: The Key to Better Natural Language Queries for NoSQL in Node.js
GOODIES TO GO! (tm)
June 21, 1999 -- Newsletter #33
Please visit http://www.htmlgoodies.com
Greetings, Weekend Silicon Warriors...
I'm back from my European vacation and it was a welcome few days off. I saw London, Paris, Lucerne, Venice, Florence, Pisa, and Rome. But now I'm back and must start writing again.
I learned that in a country where you don't speak the language, if you simply learn the words for hello, good-bye, water, toilet, and thank you, you can pretty much perform the remainder of human communication by pointing. It's true.
However, that may not be the case for much longer. CNNiN is reporting that the total number of Europeans online grew from 17.7 million in 1997 to more than 35 million in 1998 (Dataquest, part of Gartner Group Inc.).
So much has happened while I was away. An on-line drug store in Kansas sold drugs to underaged kids; a new Melissa-style virus, worm.explore.zip, popped onto the scene; computer stocks took a wicked hit; and IBM has stated that they are coming out with a mini-notebook computer that will test the fingers of normal-sized people, I'm sure. Can you tell I've been reading through news archives?
The most interesting thing about the Web and my trip was that my wife came up with what I think is a brilliant business idea. She has always wanted to be her own boss. Of the thousand or so ideas that she's run past me, this is the one we both really think will click. (Sorry, I can't tell you what it is or she'll kill me!)
I can tell you that the business will be Web-based and that she wants me to create the pages for it. Go figure. It has absolutely nothing to do with HTML or what I do. So, I've begun doing some research onto the world of Web business. Yes, I know that technically I run a Web business, but not like this. I sell advertising and give everything away for free at HTML Goodies; her idea requires no advertising banners (she refuses to have them) and involves selling a product. After close to seven years in the Web business, this is quite new to me. Luckily, I'm on summer break from teaching or I'd have no time for any of this!
The technical end of it isn't all that rough. I have no problem with hiring people to do the work for me rather than learning to do all the programming myself. In fact, I've got a couple of people who are coming back to me with bids to house the site. But what does one ask for? What do you do? How do you make the site successful?
In my search for answers I came across a great article by Dawne Shand, published on the IDG Web site. Instead of telling me what I should do, she wrote about what I shouldn't do. I wrote a piece not too long ago about the top 10 things you must do to be taken seriously as a commercial Web site. This is a list of the top 10 things you must not do. This offers a little more insight, I believe. (I added the little commentary after each item.)
10. Ignore what already exists.
There's a reason why Web businesses all look the same: They work. Reinventing the wheel will simply confuse people, so look at what the winners are doing and follow a similar path. I can do that.
9. Create psychedelic designs.
If your 13-year-old kid looks at the site and gives a long, drawn-out "Coooooooooool" then maybe you've gone a little over the top. Make it all look like business: Clean lines, white backgrounds. Stuff like that.
8. Build it to run slowly.
This is true for all Web sites in the world. Speed is what we need. My wife and I have discussed the look of the site in great detail. No frames. Limited graphics. No Java. Function over glitz. We're also thinking about paying an extra chunk of change so that we will be the only site on the server. Maybe even buying our own server. Yeah, she's really serious about this.
7. Neglect to offer a search function.
I can vouch for this one. Before Goodies had search, before EarthWeb, people were screaming for one. When I finally got one, I had to take it down because people were using it so much that it was putting a strain on the server. If you're going to run a business, people want to be able to find what they want fast. Make sure your searchable database offers multiple fields for everything you're buying. List everything that describes the product, then allow people to search (see http://www.htmlgoodies.com/tutors/database.html to make one of your own).
You know what else I've seen that works? Offering suggestions. Really. I found this strange, but apparently people like this. Set up your database so that someone could come in and say "I want to buy a present. Suggest a few things." Make it so that your database displays a few ideas for one reason or another. That's apparently very popular.
6. Confuse people at the point of purchase.
>From everything I've read, you need to make this part of the business Web site conservative and extremely easy to follow. Get it all on one page if you can. People are still very nervous about putting credit card numbers into their computer. Make it so they feel safe. If they have to go from page to page then maybe you're asking too much.
Another site said to only ask for information that is required. If you start asking people's age, height, income level, and the like, they can get nervous and maybe crash the sale.
5. Fail to acknowledge the sale.
This is equal to not saying "thank you" to someone at a sales counter. Make it so that a lavish "thank you" e-mail shows up addressed to the person. Also make sure that the e-mail contains a list of what the person bought and how much it all costs.
If you've ever bought something from Amazon.com you know that they do this beautifully.
4. Neglect back-end systems.
These are things like accounting and recordkeeping. It will probably cost a bit more, but the author suggests you integrate this right in with your order-taking rather than attempting to do it all by hand. Good idea.
3. Charge higher prices than other outlets.
A few years ago you could ask a little higher price for the convenience of buying over the Web. No longer. Know your competition. Know what they are charging and stay competitive.
2. Fail to provide a feedback mechanism.
Why wouldn't you do this? I guess it needed to be said, though. The customers need to be able to talk to you and provide feedback. And you need to provide a reply or a place where they can go to get feedback. Very important.
1. Build something your customers will refuse to use.
This is self-evident, but if your site scares people away or gives the impression that something might be shady, then people will go somewhere else or simply refuse to buy from you.
Those are some good tips that I intend to follow. Luckily, my wife will be taking care of all the money and research portions of the site. I'll just do the grunt work. In fact, once the site is up and running she wants me to teach her to code the pages so that I am totally out of the picture.
This thing's going to work. I can feel it in my bones. What I'm really looking forward to is a weekend of sales research where we go to a ton of e-commerce sites and buy stuff just to see how others are doing it.
I love buying stuff....
And that's that...
Thanks for reading. It's good to be home but hard to get used to the time change. I keep waking up at 3AM thinking it's 8AM London time. Give me a week. I'll be good as new.
Joe Burns, Ph.D.
And Remember: Do you play pool? Snooker? Billiards? Ever heard of an Englishman named Jack Carr? He became a sensation around 1825 when he made seemingly impossible shots and then placed the cue ball perfectly for the next shot. We know today that Carr put a spin on the ball we call "English." Why? Because Jack Carr, the first person to do it, was English.