Main | January 2007 »

December 26, 2006

Replacing Snap preview with Amazon Alexa

I really love the idea behind Snap Preview. I learned about it from Michael Arrington's Techcrunch - he uses it extensively to great effect on his blog. The value of a web page preview is there. I am definitely more like to click on a link if I can see a thumbnail of the page. I'm not sure how universal this feeling is, but it is absolutely true for me. BTW, I wonder why Google or Yahoo don't do this for their text ads, if it bumps click-through by a percent or two - hmmmm.....

Unfortunately, we've had to turn off Snap Preview. It was just too inconsistent for us. Every once in a while it would significantly delay the page loading. To be fair, we're changing links on the fly with Ajax and I'm not sure Snap is designed for that. We spent some time trying to decipher the Javascript but they've obfuscated it just enough where we could not tell where the problem is. In the end, since we could not fix things and decided we could not accept the occasional delay, we had to drop Snap Preview.

The good news is there is a  new alternative available. Amazon has released Alexa Site Thumbnail. We have tested it and it works quite nicely. It should be available within a week or so on Big Contacts. Two downsides. We have to pay $1 per 5,000 thumbnails (I think we can handle that!) and Amazon says there could be up to a 24-hour delay before a new site is thumbnailed (is that a word?). We will request the thumbnail as soon as someone saves a web site, so the delay may not be too noticeable. Hopefully, it does not actually take that long and/or Amazon speeds things up as the service matures.

Stay tuned ....

December 22, 2006

Features, features, features

One of the hardest things about building a product like Big Contacts is deciding what features to include. A contact manager could potentially have a nearly infinite number of features. ACT and to some extent salesforce.com have tried just that. Of course, what suffers is quality and, more so, ease of use.

Joel Spolsky, who writes one of my favorite blogs - Joel on Software, addressed this in a recent post simply titled Simplicity.  He correctly points out the fallacy of the 80/20 rule. There is a common knowledge rule that says 80% of users only use 20% of the features of a product. Therefore it would follow that focusing on those 20% of the features people would use would make for a better product. What Joel points out, and is obvious in retrospect, is that those 20% of the features are different for each person! I might use only 20% of the features of The GIMP, but there is probably one feature I cannot live without that you've never heard of.

Given that, how do you make decisions about what features to include? There are no easy answers. There are two things we try to do here at Big Contacts. One is to use the product every day, all day. We find out quickly what needs improving that way. Secondly, we listen carefully when people offer ideas. We try to find out what they are really trying to do and, if it is important, find an efficient way to do it. But, even then, we have to filter out those features that just don't seem to be widely applicable, or have an easy workaround. Like I said, it ain't easy.

December 19, 2006

Website now live

The bigcontacts.com website is now live! This is a soft launch to start getting all the pieces to start working together on the product servers in earnest.

Until BigContacts itself goes live, the coolest thing on the website right now is the link into the fully populated demo. If you want to see how BigContacts looks and works, this is an easy way to see it in all it's glory. Each time someone goes to the demo, they are get a brand-new randomly generated set of contact data. This is populated into their own instance, which is not shared with anyone. I've never seen anything quite like it and hopefully everyone will appreciate the hard work that went into creating that feature. Personally, I hate going to a demo that is empty and I have to put in my own made up data. Almost as bad is a demo full of data made up by others. What is ideal is a system that has what looks like real data to get a fell for what things will be like after using the tool for a while.

Go to www.bigcontacts.com and give it a try.