Evolvable posts

Remarkably, I work with automation and may not spend an hour a day to choose, guess from 10 titles, descriptions, or names. When you flip a coin there's 50% chance (1:1) to win, guess the side. Analogically, when you think that all of your ideas are good for others, you've got 10 good ideas for a post title or/and description that appear in organic search engine results or text ads then you've got only 10% chance (1:10) to win, the probability to guess correctly which 1 out of 10 good names or sentences attracts as most people. In fact, when it comes to too many good options then because of Analysis Paralysis,  Overchoice, Paradox of Choice, the chances aren't really so much bigger than 1:10. We know that some text can attract 10% more people and some of them even 900% more than total junk. For an example of an author, Stephen Hawking wasn't merely a genial physicist but also a book author. However, this genius often used words that didn't attract the average audience either despite his rich vocabulary (some native English speakers didn't understand him), personality, opinions, sense of humor, and his topics. His ideas were intelligent but his titles weren't the best catch for an average audience (he was too intelligent). However, you can lean on math (not the chance to win a lottery and earn on a guess). Perhaps, you don't have the time to manually check all analytics, change each one element and repeat this process 3-10 times for all (e.g. 100 posts) but if you can use automated testing of multiple texts compared against statistics of organic traffic, you can also automate the selection of the best texts and drive more traffic, you can 10 times increase your success rate of choice between 10 options, from (1:10) 10% to 100%. 

In a similar way as with testing of web design, my solution is based on an algorithm, software for automated testing of multiple titles, creating statistics (counting web visitors that came from Google (with HTTP referer) directly to land on each one of particular posts per a time period), consequently automatically, in a programmatic manner, my software finds the winner with the highest count and permanently set (uses) the most successful beast, the text that once brought, used to drive as most people. Here's how it looks in my administration panel wherein I come and write just once in (multiple titles):

You would get it readable : ))


It can be implemented in any tech environment (e.g. Python, PHP, etc). For this process, I like to play with words, personify it in Darwin's thoughts about natural selection because this is about the same. There are multiple factors (qualities of a text) that play in the game. These factors can be either synergic co-factors or actually, they can play antagonistic roles. Somebody can be an amazing source of thoughts but still, it doesn't mean that it always, all the time suits. If it's about money, I prefer to use the natural selection instead of a utopia and do it in a pragmatic - programmatic manner, with a machine that really got all the available time to perform all the tasks associated to checks of analytics and editing, I prefer the most successful choice that fits in a real-life scenario, not in a theory or New Year wishes.