Internet Entrepreneurs Blog

The term data mining has been used for many years to refer to the extraction and analysis of useful information from various sources (usually databases). Data mining practices generally overlook future growth of the data or even the value of granular data to information seekers.

With the rise of the Internet and so-called Web 2.0 applications designed to elicit user-generated content, a new analogy is useful for describing the purpose of this class of products: data ranching.

Data ranching refers to any application that exists to gather data, generally from a wide variety of individuals, and present the data in an organized and useful format. Depending on the application, varying degrees of user-input may be allowed while the output may be distributed for free or packaged for sale.

Effective data ranching requires expertise in many different areas including computer programming, user interface design, business strategy, and even human psychology. Data mining itself is a narrow aspect of data ranching involving the optimization of data presentation and analysis.

Explained another way, data mining extracts value from existing data resources while data ranching cultivates the growth of data and manages the distribution of the data product to others.

This blog will introduce the concept of data ranching by discussing well-known data ranching applications and specific businesses that are already taking advantage of data ranching today.

I recently found a little trick to save some money with my Google Adwords campaign. Supposedly Google dynamically bids on ads to get you the lowest cost per click based on your max CPC for each keyword but I wasn’t sure if this was the case.

Most of my campaigns were meeting their daily budgets so I figured instead of raising my daily budget I would try lowering my max CPC to get more clicks for the same budget. In one case I was paying an average of $0.07 per click (max bid: $0.10) on a $10 daily budget so I tried lowering my max bid to $0.07. Sure enough, over the next week my average CPC went down to $0.06 and I was STILL maxing my daily budget. Another week, another penny and today I’m down to $0.03 average CPC for a $10 daily budget. I’ve effectively increased the number of clicks I receive each day by 133% and I’m not spending a penny more!

Some might argue there are downsides to this approach since my ad positions undoubtedly fell after I lowered my bids but from what I’ve seen, the quality of the leads I receive is virtually unchanged. Take a look at your campaigns and look for those campaigns that regularly meet their daily budgets and try reducing your bid - you may be able to squeeze more clicks out of the same budget!

Good example of poor stock photo choice on the Mills Pride Cabinet website. I was looking for an assembly guide for some new kitchen cabinets I purchased and found this page:

Strange that Mills Pride has instruction guides for cabinets AND children’s bicycles (or at least that’s what I might be lead to believe based on the photo of a kid and his dad putting a bike together). I guess that’s what comes up when you search your stock photo gallery for “assembly” pics, better luck next time.

An interesting phenomenon has been brewing ever since MySpace started getting popular. Affinity group social networks are popping up EVERYWHERE and not just as an addition to already popular niche communities. I first started noticing this with websites billing themselves as “Christian MySpaces.” There are litterally dozens of these with memberships from 30K down to just a few dozen. The most disturbing thing is all these sites are apparently using the same “social networking” software (not sure which one, there must be a free/cheap one out there because EVERYONE is using the same one). I suppose the idea for many of these webmasters was that MySpace was a scary place for Christian teens and there needed to be a safer environment for blogging, meeting people, etc. But is there really a need for 2 dozen different ones?

Another group I’m certainly familiar with is the outdoors/action sports online community. One site, ActionProfiles, has created a sort of action-sports social network but with an interesting twist. Beyond simple athlete profiles and chatting, sponsors use the site to find athletes to represent their brands. Great concept but in reality most of the “athletes” on the site are 12 to 14 year old kids who think they’re the next Tony Hawk and are entitled to full sponsorship deals. Nevermind that most can’t even land a clean ollie yet, it’s all about the lifestyle.

How about creating mini-MySpaces based on ethnicity? Already being done by Community Connect and doubtless others. Is there really a need for more MySpaces, isn’t the point of the original to be a place where you can find everyone you know? Am I supposed to keep a profile on different sites for each of my interests? I’m looking forward to the MySpace bubble bursting, it already feels like it’s been going on for too long.

You can see my MySpace profile at myspace.com/singletracks

I’ve been going through all our junk in storage (shed, basement, attic) and I’ve come across some bulky collections I want to get rid of. There’s an unopened/unused George Foreman grill, a 21″ Sony Trinitron Computer Monitor, a remote “training” collar for a dog (read: shock collar), and countless other gems that should be a cinch to sell on eBay or Craig’s list.

In addition to these one off items I also have two rather large collections: CDs and Nintendos. The Nintendos span 3 generations: 2 8-bit NES systems, 1 Super NES, and the Nintendo 64 plus games for each system (50 games for the 8-bit system alone). My question: should I sell the games (and systems) as a lot or individually? Which will make the most money after accounting for listing fees?

The question is not as simple as it sounds. Listing as lots of say 5 or 10 games should give me lower selling fees on eBay but will I maximize the final selling price? Let’s start with an example. Mike Tyson’s Punch Out has been selling on eBay for between $10 and $15 on its own and the Legend of Zelda (gold cartridge) sells in approximately the same range (coincidentally I have 2 of each). If I put Zelda and Mike together in a lot of 5 games (let’s say the other games have marginal value) what should I expect the final price to be? Will the two popular games (plus the other 3) draw more potential buyers to the lot than each individual game would draw on its own? Will this send the price higher as Punch Out and Zelda fans duke it out to win the auction? Sadly I haven’t found any conclusive discussion on the subject.

I had hoped I would have duplicates of 5 games so I could set up a perfect test but alas, I do not. Looking at the values of completed auctions it doesn’t look good for lots of 5 (most sold for under $10). My approach will be to list the games individually with minimal promotion (no bold listings, no subtitle, etc.) and no reserve. For those games that don’t sell I’ll put them in a bundle to sell as a group. My plan for the CDs is similar, though I’ll use Amazon to list those instead. The only factor that may derail my plan is the time required, I’ll keep you posted on how that goes…

Man, that PayPal is some powerful stuff! I’ve been on PayPal for almost five years now (actually 4 years, 8 months, and 14 days according to my member information box) and only recently have I realized how strong PayPal really is.

Most of are probably familiar with PayPal’s eBay integration (they are, after all, owned by the same corporation) but the tools go WAY beyond simple auction payment. One of the first things many webmasters might find useful is the FREE PayPal shopping cart you can use for your website. PayPal has made great strides to allow customization of shopping cart pages and there is actually a good bit of flexibility in setting shipping charges (like flat amount, free shipping for orders over $X, etc.). Buyers don’t even need a PayPal account to buy through your cart, they can simply use a credit card just as they would on any other e-commerce site. Best of all, you only pay processing fees when purchases are completed as opposed to services like Authorize.net that charge multiple minimum monthly fees.

Another awesome feature that some might not be aware of is Instant Payment Notification (IPN). IPN is a great way to sell online subscriptions (we use this on singletracks) because it automatically notifies your website when a successful payment has been made. Integrating IPN is a bit tricky but there is great support online and PayPal offers great tools for testing your IPN system before you go live. If you offer instant access to online content via PayPal you may want to disable e-check payments since these take a couple days to clear which can be confusing to some users.

My latest PayPal discovery (and I’m actually pretty behind on this) is the ability to purchase and print USPS postage online. We’ve been using Edicia’s Dazzle program for Unity3 and it is a huge time and money saver but it’s kinda hard to get over the $15 a month fee we pay to use the service. With PayPal you can print shipping labels through the site without any special software and no subscription fee! Delivery confirmation is included and you can even purchase insurance through the site. Best of all, it automatically records and notifies your buyers (assuming they paid through PayPal) and even disguises the actual shipping cost on the label. My one complaint is that the label printout takes up an entire sheet of 8.5″x11″ paper while the actual postage label is only half a sheet (labels are expensive). Apparently people have found ingenious ways to save labels like feeding half sheets of paper into the printer so there are ways to conserve labels if you try.

PayPal is also a legit bank and I haven’t used their ATM/Debit Card or banking services but it seems like a good idea if you use your account alot. If you’re looking for a good way to start selling online, consider using PayPal for your shopping cart and payment gateway. How can you argue with free?

I really wanted to like osCommerce. I mean, it’s one of the most popular shopping carts in use online and has a huge developer community building modules to do pretty much everything you can imagine. Plus it’s open source so you can download a copy for free, a huge deal when you compare to the prices of most other carts. Since you have access to the source you can customize EVERYTHING and there are several companies that are doing just that to great effect. I had a couple weeks to get my e-commerce site up and running and I was getting very frustrated with my inability to customize my Volusion website so I decided to give osCommerce a try.

One of the first hassles I ran into with osCommerce was trying to figure out how to do a secure install with my webhost. I don’t really know much about SSL and certificates and unfortunately my hosting company was of little help in this regard. I did manage to move my hosting account to the “secure server” (whatever that meant) and was ready to install.

The osCommerce software package out of the box has most of the features you need for an e-commerce site: product catalog, product categories, shopping cart, etc. However, I also needed the ability to create coupons and promotions, sell gift cards, and track inventory at the option level (i.e. t-shirts by size and color). Luckily there are some highly recommended “mods” available to do just these things. Perfect!

I had a little trouble following the detailed instructions required for installation, luckily BBEdit’s compare function saved the day. You see, to install most osCommerce mods you need to edit the source code for multiple PHP files by hand just to get them working. Missed a line of code or copied it incorrectly? Too bad, you probably just broke your shopping cart. Fortunately there is a huge community of volunteers out there willing and able to give support via the osCommerce community forums.

osCommerce has been around for several years (more than 6 years now) and there are more than 10K online shops using osCommerce today. Many of the software mods and add-ons have been around for almost as long, meaning you’ll need to catch up on a few years’ worth of postings before you ask your question on the forums (etiquette, remember?). Chances are someone else had the same problem or question as you but it’s nearly impossible to wade through all the back-and-forth contained in the forums. Then again, maybe you’re the first to encounter such a problem…

Since there are so many mods and add-ons available for osCommerce and there isn’t a central body coordinating all the add-ons, you’ll probably run into some compatability issues when you try to install more than one mod. With more than 2,000 mods available, you can do the math and figure out how many possible unique combinations one could come up with for their installation. I just needed 2 modules: gift card and voucher system + an inventory system. Even with just 2 mods I ran into issues trying to get the two to play nicely together. After 2 days of screwing around, I was done.

Don’t get me wrong, I’m not an amateur hacker hoping for a double-click install here. I just need something that I can see could potentially work. I was also quickly frustrated with the administrative back-end which paled in comparison to the commercial packages I considered. I never even made it to the point of customizing the look of my store but I imagine I would have run into just as many frustrations.

If you’re a web design and coding firm it might make sense for your team to take a few months to build a scalable working version of osCommerce with a robust templating system and the most popular e-commerce features built in (like Varien did). If you’re a small online business operator, stick with something cheap like Yahoo! Stores or even free like Big Cartel and save yourself the effort.

A few months back Joe and I started getting friend requests from strange women on MySpace. Now this in itself is not interesting or really all that unusual except for the fact that these messages would usually hit just after logging in to MySpace. Essentially if one of us logged in to check new messages or something, we would be hit with up to a dozen friend requests in the space of the 5 minutes we were online. It was as if the spammers had an inside track on who was online when. Not only that, on at least one occasion Joe clicked on a user’s name to view her profile and was quickly wisked away to a (non-Myspace) adult content site. Had the spammers taken over MySpace?

At the same time we also noticed spam coming through our Skype phone. We set up a Skype account for our business to make free long distance calls using the screen name messengerapparel (our company name). Now, several times a day, we get messages from India, China, the Phillipines, and Turkey offering us information on garment production overseas. The Skype spam is especially annoying because we somehow get pulled into random apparel chatrooms, requests for our contact info, and messages written in broken English all on our business phone. There must be some way to turn all this off but one would think the default configuration would keep the spam out a little better. How do you even search for random users on Skype, these guys must be pros!

It seems even the newest mediums of communication and expression quickly are hijacked by spammers for their own use. I would say mobile phones are next but in fact, that’s already been done (anyone else get the snakes on a plane text message?) If you think email spam is a big problem think again: spam will find you where ever you are!

You have no doubt been waiting with bated breath for the conclusion to my e-commerce article to find out which solution I chose for my first real e-commerce project. In the end Volusion seemed to meet most of our needs for a reasonable price. But despite having the opportunity to try the software free of charge for 2 weeks, there were still surprises and disappointments along the way. Nothing is perfect, right?

First off, the ability to customize Volusion stores is much more limited than I had hoped it would be. Much of the shopping cart, product pages, and category pages are locked up in ASP files that can’t be touched and in many ways you’re just able to frame the Volusion layout with your own template. Great for people without the time or skills to edit HTML, frustrating for those who do. Support for CSS is spotty at best, though with better penetration it could be used to really allow full customization. To make most of the design changes you need to edit/replace the stock background, button, and layout images, a time consuming and frustrating process.

I initially signed up for the cheapest plan Volusion offers thinking it would be more than enough to get started (100 products, 1Gb of data per month, 50 Mb storage) but quickly found that Volusion’s definition of a product is different than most. After adding just 25 items (t-shirts, four sizes of each SKU) I had hit the 100 product limit. It turns out “product” means SKU/option combination when using option-level inventory control, forcing us to upgrade to a pricier plan just to get our initial inventory loaded (less than 60 SKUs). Frustrating, and seemingly a little deceptive.

Technical support has been good with prompt and courteous responses to most of my inquiries. In one case I was told something was not possible just as I figured out how to do it on my own. No worries though, this is what happens when you’re a technical genius like I am

The Volusion servers seem to be a bit slow at times and the file caching they use is pretty aggressive (good for bandwidth I guess, not so good for making quick changes).

Although my comments may seem overwhelmingly negative, I’m really quite happy with the choice thus far. The payment integration was simple and the backend admin area is powerful (especially compared to my osCommerce foray). The SEO tools seem solid, though our site isn’t listed in the search engines yet so it’s tough to tell. The data system seems to be very flexible and exporting info to Froogle and Endicia (for shipping) has been a snap. I’ve even been able to build custom ASP pages with database hooks (though not officially supported) that work like a charm. The support forums are great and other Volusion store owners are more than willing to offer advice and support.
Volusion really seems to be the best hosted e-commerce solution around, despite the limited customization options currently available. I’m looking forward to more e-commerce solutions that take things to the next level of customization and powerful back end support (keep an eye on these guys).

I’ve come to realize a major shortcoming of Google AdWords (and similar search marketing options) is the inability to advertise innovative products and services. What do I mean?

Well suppose you just invented a electric widget. No one knows about the electric widget yet or the benefits of owning such a device and so they certainly aren’t searching for “electric widgets” on google or anywhere else. You could try to buy search terms related to your competition and build ads saying something like “Still using a manual widget, try the electric version!”

But what about if there isn’t even another thing like the widget out there? You could try to buy keywords that your target market are searching for (suppose many people in the market for electric widgets also enjoy playing soccer). But if you buy “soccer” as your keyword your ad will quickly disappear due to its irrelevance to searchers’ intended information.

Let me make this more concrete. I’m working with a partner on launching a site called Unity3 that sells “inspired fashion” clothing (basically, clothes for Christian teens and twenty-somethings). There is some search traffic for “Christian clothes” and the like but for most Christian teens there is very little awareness of the Christian clothing category at all. Our hope is that the market will continue to get larger (due in part to our marketing) but this isn’t possible through advertising on obvious terms.

One of our strategies has been to advertise alongside other keywords these kids ARE searching for like Christian music (bands like Switchfoot, Anberlin, etc.) The search traffic here is HUGE compared to the Christian clothing traffic and the demographic is right. However, since our ads aren’t as tightly targeted as, say, a CD retailer, our ads are quickly deactivated for search.

What does this mean? It means online retailers in newly or unestablished markets can’t rely on search marketing alone - mainly because no one knows what to search for yet! New markets require old school marketing - print, television, etc. - which can be expensive. So while first mover advantage is key, it is also costly.