4 ms·
I have experimented with this, and to my joy, it works. In first attempt at rhythm, I emailed users on a consistent basis. Soon I discovered that they were gett
by chewxy 14y ago
I have experimented with this, and to my joy, it works. In first attempt at rhythm, I emailed users on a consistent basis. Soon I discovered that they were getting ad-blindness (well, sorta). So now, my new version emails them once every 2, 3, 5, 8, 13, or 21 days, with the probability of a user being emailed weighted over the slower end
- saraid216 14y ago> with the probability of a user being emailed weighted over the slower end This doesn't parse for me. What does it mean?
- chewxy 14y agol = [1,2,3,5,8,13,21,34] <--- this is the list of fibonacci numbers. The probability of a user being on the track of once every l[7] is higher than the probability of the user being on track for once every l[0]. Users change tracks on a monthly basis. So if you're on a once every 21 day email cycle for June, you might be on a once every 8 day email cycle for July, although the probability is high that you will stay on the once every 21 day cycle
- saraid216 14y agoInteresting. Thank you.