In terms of income conformance I would describe the second half of July as an abysmal month punctuated with occasional goodness. Before the switch to Datayze on July 9th, I made $11.52 or $1.44 a day. After the switch I made $17.53 or 79.7 cents a day. A bit more than half the previous daily rate. During the second half of July I had 3 days under 15 cents, and another 6 under 30. Overall it was better than I feared, but the dark cloud hanging over August is that there’s no clear uptick at the end of the July, so I can expect this ~80 cents a day rate to continue this month.

That’s the bad news. The bounce rate is still under 3%, pages per session were 4.43 and time on site is up to 2.45 seconds. The first two numbers are not as good as last month, but within striking distance and not terribly unexpected after a re-branding.

The issues I’ve been facing with Search Engine Optimization (SEO) post the move inspired three new apps.

Site Navigational Analyzer. This one I built when the Google and Bing search engines were slow to index my site. I was curious how interconnected everything actually was on my site, and how varied the anchor text is. The intuition behind varying anchor text is that it can provide different hooks to help a user find where he or she wants to go. Branded links (links with the name like “Site Navigational Analyzer”) are great, but descriptive links also help users who don’t recognize the title, or for where the title is too short to include a site’s full capability or article topics covered. The “switch to Datayze” link above is a good example of a descriptive link.

Site Validator. This is another website Spider app that crawls through a domain and creates a report of errors sorted by how common they are, as well as a report for individual pages. I used this to fix not only errors on Datayze, but on sktyler.com.

Finally I created a Thin Content Checker. Thin content is SEO nomenclature for “with little value.” While content value and page length are not the same, they are somewhat linked. The industry standard amount SEO experts is that under 200 words for most applications should be considered thin. I designed my thin content checker to be flexible enough that it can ignore headers, footers, & the like, as well as tell you the unique content per page.

The drawback with spiders is that they are a bit more resource intensive than most of my other apps. The Same Origin Policy prevents me from implementing the spider in Javascript, and moving some of the resource burden to the User’s machine. For now I opted to cap my spider at 400 pages per day, which is larger than most personal websites, but I’m also considering creating a teared user model which would allow some users to have an uncapped rate.

My plan for August is to keep focusing on the numbers and see if I can’t bring them back up.

Posted in Work Life | Tags: Sarahsoft