Other Data FAQ
Why the same ad network has different fill rates for different products and different PIDs?
It is normal for the same ad network to have different fill rates for different PID ad slots of different products.
The fill rate depends on the resources of the ad network, product quality, user quality, floor price setting, users' conversion effect on the ad, etc., so the fill rate varies.
What determines the ad fill time?
The ad fill time is determined by the server of the ad network and the users' network conditions.
- Due to the server of each ad network , the fill time of each platform is different.
- The fill time of each region is different. For example, the network condition of the T1 region is generally better than that of the T3 region, so the fill time of the T1 region will be faster.
- In most cases, higher-priced ad sources take longer to fill than lower-priced/unpriced ad sources.
Why is the show rate low?
| How to analyze | Explanation |
|---|---|
| Whether the ad scene is too deep | If the ad scene is too deep, or the ad trigger conditions are too strict, the ad show rate will be low |
| Whether the app limit impressions | If limit the number of ad impressions and opportunities, the ad show rate will be low |
| Whether there are invalid impressions | If there are invalid impressions, the ad show rate will be low |
| Whether the number of ad cached is too much | If the number of ad cached is too much, some cached ads will not be showed, the ad show rate will be low |
| Whether the ad fill time is too long | If the ad fill time of the app is too long, although there is ad fill, the ad will not be showed in time, the ad show rate will be low |
Why is the number of ad fills always much greater than the number of ad impressions?
Show rate = IMPs / Ad Fill*100%. Ad fill is at the app level. As long as the user opens the app, a request will be made to all ad sources (because of the function of ad preloading), and then the ad sources will be filled. Therefore, the number of ad fill will be affected by the number of ad networks and the number of PIDs.
However, the ad show needs to be triggered by the user. The user must enter the adscene and trigger the ad, and then the ad will be showed. So the number of ad fills in the ad source is always much greater than the number of ad impressions.
Why does the show rate of a product in the performance report exceeds 100%?
Show rate = IMPs (of ad networks) / Ad Fill (of ad networks)*100%
Since some ad networks do not provide or return ad fill numbers, there may be cases where the number of ad impressions is greater than the number of ad fills, that is, the ad show rate exceeds 100%.
Ad networks that do not provide or return ad fill numbers: Unity/Vungle/Applovin/Mopub/ Pangle/myTarget/Tiktok Ads/Kuaishou/Sigmob, etc.
In the waterfall, why is there a situation where the impressions of low-priced ad sources are higher than those of high-priced ad sources?
The number of impressions of an ad source is related to both the ad fill rate and eCPM, not only by eCPM.
To give an extreme example, the app only configures two ad sources A and B, the fill rate of A is 1%, while that of B is 100%, and the eCPM of A is higher than that of B.
The app has 100 ad show opportunities, and the ad source A successfully fills 1 time, and gets 1 impresssion because of the higher price;
And ad source B is filled every time, except for 1 impression opportunity obtained by A, the rest are showed, a total of 99 times.
It can be seen that the lower-priced ad source B has more impressions due to its higher fill rate.
Does the higher the eCPM of the bidding ad source, the more impressions?
The number of impressions of bidding ad sources is related to both the bidding response rate and eCPM, not only by eCPM.
The bid response rate is the rate at which the bidding ad source has a response after receiving the bid request.
If there is a bid response from the bidding ad source, the ad can generally be requested, so the ad fill rate of the bidding ad source is very high, basically above 90%.
For bidding ad sources, the bidding response rate can reflect the actual filling situation.
To give an extreme example, the app only configures two bidding ad sources A and B. The eCPM price of A is higher than that of B, but the bidding response rate of A is 1%, while that of B is 100%.
Suppose the app has 100 ad impressions. Ad source A gets 1 impression; while ad source B responds to every auction, except for the first impression opportunity, which is acquired by A. B will get 99 impressions. It can be seen that the ad source B with lower eCPM gets more impressions because of the higher bidding response rate.
Why some ad networks have lower eCPM after switching from waterfall to bidding?
In traditional waterfalls, if an ad network configures a high floor price or has a small number of ad sources, there will be fewer impression opportunities.
For example, if an ad network only configures an ad source with a floor price of $50 in waterfall, the ad network will only fill ads with a minimum price of $50, so the overall eCPM performance of the ad network is $50 or more.
After switching to bidding, if the bidding floor price is not configured, ads at all prices of the ad network are accepted by default, and both ads above $50 and ads below $50 are displayed, so the overall eCPM performance will be lower than $50.
Does low ad show rate affect revenue?
The show rate is affected by the request strategy, the repeated request strategy, the number of ad networks, and the number of PIDs. A drop in show rate doesn't necessarily lead to a drop in IMP / DAU.
However, when the show rate is too low, attention needs to be paid to it, as it may be judged as low-quality traffic by the ad networks.
What are the factors that affect the CTR?
| How to analyze | Explanation |
|---|---|
| Ad scene design | Is the ad scene design easy to cause accidental clicks |
| Check the delivery end | Whether the user quality is poor |
Is CTR directly linked to eCPM?
There is a certain correlation between CTP and eCPM, but not the higher the better.
eCPM is determined by the ad conversion rate of the ad networks.
What determines the eCPM？
The price of eCPM depends on whether the ad resources of the ad networks match the traffic, and on the judgment of the ad networks on user value.
Why overseas eCpm is usually lower than domestic eCpm for the same ad unit of the same product?
Generally speaking, the competition in overseas markets is more intense, so it is normal for most products to have this data feature.
Will the waterfall flow too long and cause the ad to return slower?
If the waterfall flow is too long, it will affect the return time. It is recommended to decide whether to optimize the length of the waterfall flow according to the ad ready rate.
If the ad ready rate decreases, you can appropriately reduce the number of waterfall layers, or place the ad source with a shorter return time at the head of the waterfall.