It was 15 years ago today that I "programmed" my first Internet-based survey, while working for a local marketing research call center. I think about how far this business has come since then, and I'm simply astounded.

The hot new Web browser "everybody" was using was Netscape Navigator, which was a huge leap forward from Mosiac at the time. I put "everybody" in quotes because very few people had Web access. In fact, to get Web access, I had to go to Computer City and buy "Internet in a box", which was a box of freeware that sold for about $30! Since Web penetration was so low, we pulled this first survey off using plain text Email messages. Not Email invitations to Web surveys, but the Email message was the survey!

To our delight, we got a tremendous response rate on that first project. Almost 95%!

The idea was the respondents were to hit the their Email clients' (Eudora and Pegasus, mostly) reply button, and type Xs into the boxes that represented their responses. Since this was so new to the world, no software for extracting the answers existed... At least that I was aware of. So we printed out every response and performed manual data entry on the whole lot of them.

We actually learned a lot about online data collection during that first survey. For example, we learned that, given the opporrtunity to do so, respondents will take it upon themselves to add and alter survey response choices. How one can quantify those answers, I don't know.

It wasn't long after our first (and only) Email survey that we took this practice to the World Wide Web, and the rest is history.

About a month ago W3 programmed a survey for that first client that hired us to do the Email survey. I offer my deepest heartfelt gratitude for the years we've spent working together. Your business and your friendship are greatly appreciated.

I colleage once told me that everyone should redesign their Web site every few years, whether it needs it or not. Well, today was the day that I had a few hours to spare, and decided to throw a new design together. As much as I don't like the term "Web 2.0", the idea was to emulate that look.

So, it's out with the old and in with the new. I hope you enjoy it.

I received an unsolicited commercial Email message (spam) from a guy named Tim Roberts today... Twice.

The messages were virtually identical, with only a slight variance between the two. The second one extended the special sale price by one day. It may have had a few other tweaks, but I didn't read it that closely.

The message was advertising an Email list of 4,000 market researchers.

When people spam me with offers of cheap life-enhancing pharmaceuticals and various sketchy services I shrug them off. When companies start dragging the market researcher profession into their world, I get furious, and I do something about it. After reporting this gem to SpamCop, I decided to do a little research on this company. I went to the company's site and read up on their offerings. This, of course, led me to their privacy policy.

While they didn't specifically mention the word "opt-in" to describe any of their lists, they certainly alluded to it a few times, suggesting that members "provide us with your email". Beyond the shadow of a doubt, I can assure you that I am not a "member" and I have had no previous contact with the offending company. I have relatives I won't give my Email address to. I certainly did not give it to them.

This experience got me wondering about this practice, and if people fall for this sort of thing. I'll write more about this another time, but let's just say (for those of you who are still new to the Web survey game) that your online reputation is probably more important than you realize.

So, if you're contacted about lists by a company you've never given your Email address to, do you consider the source? In other words, if they're willing to spam you for your business, are you going to trust for a second that they built their list/panel legitimately? For your sake, I hope not.

It's official. W3 Surveys has sold out to pop culture and joined the masses on the social networking site, Twitter.

To follow us, please visit our Twitter page at http://twitter.com/w3surveys.

So, what are you going to get from us on Twitter? Obviously we'll use Twitter as a way to broadcast corporate communications, such as letting you know when we update this blog, but we hope to take it a little further... To an interesting, but less research-oriented level. We're pretty serious nerds here, and we like to talk about the latest in technology. So don't be surprised if we use Twitter to rev you up about some cool sites, gadgets and gizmos to help you in your daily life.

Does your online market research provider charge you a cost per interview (CPI) for each completed survey? If so, next time you call for a bid, ask them why.

I've been astounded by this practice since Web-based surveys first hit the scene. I have to assume that they were thinking along the lines of "we've always charged them a CPI for telephone, mall, door-to-door and mail surveys, why not charge them one for this new medium also?" Well, I'll tell ya why not: There are no data collection costs. There is no labor, long distance and no postage. Your field service programs the questionnaire, sends the invitations and goes out for a nice lunch on your CPI.

Sometimes your provider is not the one to blame. Many of the Web survey software companies charge a price per complete simply because they can. If your field service has to pay it, you have to pay it. Though I've never understood why a software maker felt a need to charge us per complete. Especially if we are hosting it ourselves. I suppose it's just another opportunity to charge their customers for something.

Now, there are exceptions to this. If you are using a panel (theirs or a 3rd party's), you should fully expect to pay a CPI. They are either paying money to a sampling provider or are paying incentives to their own panel members. Not to mention the high cost of maintaining and replenishing a robust panel.

Also, if you are conducting a very high number of interviewers (5,000+). Bandwidth does cost money, but you shouldn't expect to pay for it directly if you're doing 100-500 interviews.

If you're not conducting a large number of interviews or using a panel, stand up for your rights and tell your programmers that you do not feel you should be paying a CPI.