After struggling with 100+ orders per day on ebay, I realized I needed an automated system. Ebay and paypal are notoriously difficult to work with for large scale operations and the applications that interface with them tend to be short on features. With this in mind, I set out to create my own order processing system.

Before Paypal had an API, it was very difficult to get buyer information for creating internet based shipping labels. This program allowed you to log into your account and retrieve shipping information for your buyers. It also saved transaction snapshots, so you wouldn’t need an internet connection to view archived sales.

This application allowed you to keep in constant communication with your buyers to ensure they were happy by using a click and email system that notifies buyers of their shipment’s status. The notification idea is based on the system used by Dell for shipments where the customer is notified of shipment status through frequent, step-based email.

For example, a step 1 email would say order received, and would be marked as step 1 of 5. A step 2 email would say order confirmed and so on. For my order system, I used the following steps: order received, order processing, order processed, shipping label created, and order shipped. Using this method, customers were much more likely to leave positive feedback because I had kept in constant communication with them through a simple point and click interface.

Once Paypal created an API, I moved to Paypal Helper 2.0, which takes advantage of the new API to streamline large scale ebay order processing.

Dell ran a Christmas themed contest a few years ago (2004-2005?) called the Dellf Sweepstakes. The idea behind it was pretty neat at the time: register your email address, fill out a CAPTCHA box, and you get three chances to win prizes.

The CAPTCHA design made it very easy to reverse engineer (an evenly spaced grid, solid background, unaltered text). By using the photoshop API and a custom written action I was able to take this:

and turn it into this:

From there, I sent the edited image to the Office API and did a quick OCR command to get the text from the image. Then, using the WebBrowser control, I scripted the actions to put the text into the web page and register an email address.

Here’s a view of the program in action:

It’s important to note that I didn’t collect any prizes. This project was more about the thrill of the chase and not so much the actual prizes. I ended up “winning” almost every prize including the grand prize, a 42in. plasma TV.