Personal sorting notes, there is subjective judgment.
Several noteworthy points that affect weights
- downloads
- View/download the conversion
- Expression of related words
- Elimination of untrustworthy factors
- Long-term stability
Downloads are easy to understand. Without enough downloads, conversion rates have no credibility. Assuming that a product has 10 views and 1 download under a certain keyword, it is certainly not enough to give a right. But if it’s 1,000 views and 100 downloads, it’s a whole different story.
Conversion rates can only be talked about if there are enough downloads, but from a developer’s perspective, conversion rates are the most important thing.
It is worth noting that the keywords of the product are not isolated, and have auxiliary reference effect on each other.
Another thing to watch out for is the reliability of the data, which means Apple will weed out untrustworthy fakes, usually by looking at the authenticity of the device, the volatility of the data, etc. (mainly to combat the top scorer).
Apple watches the numbers change over a period of time in order to see how products actually perform, and only products that perform consistently enough will have the opportunity to give promotions.
Change period of weight in time dimension
- New stage
- Initial weights
- Stable weight
- To review
- Loop to step 2
The new product phase is easy to understand. First of all, the App Store wants to measure the value of a product by the two important indicators of download and conversion rate, but unless users buy and release, there is no reference for conversion rate. So when the product was first released. Apple gives a “rookie bonus period” of sorts. At this stage, a relatively high amount of display will be given through the relative top position of the keyword ranking, but this amount will not be very much, and eventually about thousands of display a day.
The next three to five days are the preliminary weighting phase, which sums up a relatively stable download rate that is Apple’s “first impression” of a product. It’s the same with dating. Whatever the other factors, once a first impression is made, it’s hard to change.
Many product launches are “three days of life and death”. The number of downloads is almost the same on the day of launch, but the changes over the three days are quite different, with some adding more and more, and some adding less and less, which is the result of the weighting factor.
If it does well, give you more, if it does better, give you more, and so on, and vice versa. Until the performance is found to “match” the corresponding position, the product enters a stable phase of weight.
Because the App Store ranking rules change, or there are times when a product has made a major update (either in form or keywords) that Apple thinks it needs to be re-evaluated. It is possible to reset the weight of the product and see how the conversion rate changes based on the product’s performance. This is equivalent to originally the other side of your appearance level impression is quite good, you suddenly changed the model greatly, the somebody else can not re-evaluate you.
So sometimes when the product performance is good, don’t wave, wretched steady development to observe the market performance. If the quality of the product itself can stand the test of reset is good, if it can not stand to be reset, it is difficult to go up. As the saying goes, what luck wins, skill loses.
Summary of product release and update strategy under weight rules
- The first version of the product can be simple, but the display art and App Store page should not be arbitrary.
- There are a number of reasons not to ship a second release quickly after a release if you are not in a hurry to fix the problem, and one related to this discussion is that it undermines the reliability of data observations.
- Avoid large random replacement of keywords, as far as possible planned gradual change, and the need for a cycle to observe the effect, which will bring high uncertainty of weight.
- If the first release doesn’t work out and the conversion rate isn’t high, decide early on whether to continue or not, and if you continue, emotionally accept a long period of low growth ahead of time.
Related portal
ASO experience – App Store submission and fill in details
ASO experience (1) keyword selection
Mobile Developer Alliance Membership guidelines