NotWise
Desert Rat
- Joined
- Sep 7, 2015
- Posts
- 15,927
I know there were some comments in this thread about the new list. I have some anecdotal evidence.
My romance story fell off the first page on day two. After two full days I had 5500 reads. On day three I only had about a hundred reads.
The main Valentine listing page didn’t seem to help much.
I'd love to be able to tell whether my readers are coming from the New list, the category hub, the story page, or somewhere else.
I have a simple conceptual model for how views vary over time. Basically, the reading population is sampled without replacement, and there's a different population for each page linked to the story.
My Summer Lovin' contest had an unusual (in my experience) pattern over time, and I was able to use my simple conceptual model to match the pattern very closely.
I could tweak things to get closer, but I don't see the point.
The two doglegs on the curve correspond to when the story fell off the I/T hub and when the contest page was unlinked.
I can't distinguish the effect of the New list from the I/T hub. I think that's because the I/T readership overlaps so much with the general readership. The story dropped off the I/T hub after two days and at that time there was still a lot of interest in the story. Cutting off the readers on the hub before the population was done with the story caused the dog leg.
The second dog leg seems to have a similar cause. The contest page was unlinked on October first when the story was still getting a significant number of reads from the page. Since then, the story has had about 84 views per day from various sources.
My simple model was that the category hub and/or New list gave me a potential viewer population of 45,000, and the probability of them viewing the story was high -- 87% per viewer per day. The potential viewer population for the contest page was 40,000 (almost as high), but the probability of them reading my story was only 3.5% per viewer per day. The last, straight-line leg indicated a large population, but only a very small probability per day that they'd view the story.
So in this case, the contest page seemed to have a big effect. It's only the doglegs that make it possible to take a guess at what's going on, and I don't usually see the doglegs.
