The Terms of Service has been updated. Please go and read it.
The content of the ToS has been shuffled and reorganized with headings. Some sentences were edited for brevity, and in a few places edited to remove duplication. No new clauses or stipulations have been added, so there’s no need for anyone to re-agree to it; if you are a seller who already agreed to the ToS, you’re still OK
We’ve added some video tutorials to the site, showing how to accomplish common tasks, like using the shopping cart, searching for products, checking out, and retrieving download tokens. These initial few are just the first 7 out of a planned series of 29 topics that will be produced as video tutorials in the next few months. As more of them are produced we’ll announce them here.
To see the tutorials, go here:
http://www.scubbly.com/help/tutorials/
You’ve seen it already on other e-commerce platforms; anonymized purchase data is used to generate suggestions. Just last night, we added this feature to Scubbly: when you view product details, below the product description you might also see a section titled “People who bought this, also bought…” with a list of up to 5 other products that have been purchased by the same buyer.
Of course, this only happens if a buyer purchased more than one thing, and this was one of them. Since the overwhelming majority of sales at Scubbly are one-off purchases, not very many products have any stored association.
How it works
It’s powered by a simple “association matrix”. Product A might be associated with Product B, and that association has a “strength” or “score”. Associations between products can be visualized as a 2-dimensional grid or matrix, where the number of potential associations between products is ((n^2)/2 – n), where n is the number of products at Scubbly. With over 50,000 products currently being sold at Scubbly, that number is an astronomical 1.25 billion. If two products are associated, the score is greater than zero. The more times a product is purchased with another product, the higher the associated score between them.
Using an ingenious algorithm invented by Scubbly, these suggestions are compiled in real-time. The moment a purchase is made, its anonymized details are propagated into the association matrix. It sounds a lot more complicated than it really is.
Enjoy!
Hi everyone!
The Feb 1 payout just went out moments ago. Congratulations to all the sellers who achieved impressive numbers this month.
January was an eventful month, for sure… Most notably the massive hardware problems mid-month. After a sleepless 40 hours of restoring the basic service, there were days of forensic accounting, laboriously reconstructing “reconcile” transactions from the raw PayPal transaction logs… In total, the site was down for nearly 20 hours, and only partially in service for about 5 of those. Big thanks again to the crew at RackSpace for getting us set up on the new servers relatively quick.
And yet, despite the downtime, January was Scubbly’s highest payout month ever, with more sellers edging over the $50 payout threshold than any previous month!
What lesson did we learn? DON’T keep your all your data backups in the same server farm as the databases you’re backing up. If something happens to wipe out the whole rack, well… Let’s just say that’s not a lesson we’ll need to learn twice.
Now we have “paranoid level” backup strategies in place: daily Linux imaging, along with frequent data dumps being copied out to a remote & secure storage service. We also have “panic logs” being replicated out from the main server to machines here in the office. We’re also budgeting to add some redundant server resources so our databases will form a cluster array, guaranteeing greater stability and failure-resistance. That will likely be set up some time in 2013.
Now we’re back to working on some new features. Here’s some of what you’ll see in 2012:
If you have any suggestions how to make Scubbly better, drop us a note.
Cheers
© 2012 Scubbly.com
