CaddieOpen in CaddieDownload for Mac

Do This to Increase Your AdMob Show Rate

Shared by Caddie Library · Updated
App Masters12:37Open on YouTube
Chapters show their main points

The bottom line

Raise your AdMob show rate by finding where ads fail to display, then fixing that cause one test at a time, but only when the rate is below 50%. Show rate is the share of matched ads that actually appear in your app, so each fix adds ads shown and revenue.

Key takeaways

  • Show rate is the share of ads Google returns that your app actually displays. 1:00
    • Of 20 ad requests, 10 come back filled, so the match rate is 50%.
    • If only 2 of those 10 are shown, the show rate is 20%.
  • Only a show rate below 50% is worth working on. 2:54
    • Google gives no official target. These bands are Sid Gupta's own rule of thumb.
    • Under 30% is low and worth fixing now. From 30 to 50% is mediocre, so test when time allows. Above 50% is very good and rarely improves.
  • A step-by-step drill-down finds where the show rate drops. 4:45
    • First check the integration with Google's ad inspector and mediation test tool.
    • Then break the network report down by format, platform, app version, ad type and ad source.
    • Then read the ad request error codes with listeners, and run A/B tests in Firebase.
  • Most low show rates come from a few common causes. 8:08
    • Ad sizes that don't fit some devices can be fixed with adaptive banners.
    • Ads placed on screens users rarely reach, or loaded too late because users move on fast, never get shown.
    • Preloaded ads can go stale, so check the is-ready function before showing, and slow native and banner refreshes.
  • Patience and many experiments are what finally raise the rate. 7:42
    • The process is long and tedious, and most developers need a good amount of time to reach a fix.
    • A test can give a read with as few as 10,000 ad requests.

Match rate is the total number of ads that Google has for your inventory. Show rate is the total number of ads that you actually showed in your app.

Sid Gupta

Chapter summaries

0:00 Show rate and match rate are different problems

Developers often mix up show rate and match rate, but only show rate is the lever here.

  • Show rate and match rate are the two rates to watch besides eCPM and CTR. 0:54
    • Match rate is a separate problem and is left aside.
    • Showing more ads fills more placements and earns more.
    • Steve calls show rate almost like the main component of ad revenue.
    • Sid says many developers confuse the two terms.

2:44 When a low show rate is worth your time

Whether to worry depends on your current rate, because a low one drags down CPMs and revenue.

  • Fixing a low show rate lifts CPMs and revenue, so its priority follows how low it is. 3:14
    • The answer to "should I worry?" is "yes and no".
    • Sid says please don't quote me on it, since the bands are only his experience.
    • From 30 to 50%, run experiments for a long time when nothing else is in the pipeline.
    • Above 50%, other ways to raise revenue beat chasing the show rate.

4:36 Finding where the show rate drops

Rule out a broken integration first, then use the network report to narrow down the cause.

  • Check the integration first, because an update can silently break it. 4:54
    • Add a test device in the ad inspector and the mediation test tool.
    • Sid admits it is obvious but says developers sometimes goof it up during an update.
  • The new network report shows which slice of traffic has the low rate. 5:23
    • Google added new parameters to the ads activity report.
    • A single bad build can drag the rate down, so compare version numbers.
    • Splitting by ad type can expose a sizing issue, where returned ads don't render for one image or ad type.

6:19 Error codes and Firebase tests

Error codes show why a request failed, and Firebase tests show whether a change fixes it.

  • Error codes reveal the cause of a failed ad request. 6:46
    • Google answers each request with an ad or an error code.
    • Listeners return the code, which separates a configuration fault from an ad delivery fault.
    • There are about 15 or 16 codes, and each means something different.
    • Other partners in mediation may fail to render ads for many reasons, so check the ad source first.
  • Firebase A/B tests check each fix. 7:20
    • Test whether a weak placement can be replaced.
    • If one version is the problem, try disabling ads in it.
    • Firebase is assumed to be integrated already for other reasons.
    • Sid Gupta calls Firebase the answer for running these tests.

7:59 Size and placement causes

Ads often fail because they don't fit the device or sit where users never go.

  • Ad sizes that don't fit some devices block rendering. 8:16
    • A banner or native ad in a fixed size can lack space on certain devices.
    • Switching a banner to an adaptive banner fixes many of these cases.
    • Adaptive banners change width and height automatically to fit more sizes.
    • First check whether one app version performs badly.
  • Ads on screens users skip or rush past never show. 9:09
    • In a hard game most players quit after level three, so ads on levels four and five load but never appear.
    • Fast users leave before the request returns.
    • Lowering the difficulty can bring more players to those ad screens.
    • The ad was requested but is not loaded when the next screen opens.

9:48 Devices, caching and refresh

Weak devices can't be fixed, but stale cached ads and fast refreshes can.

  • Low-end devices are not worth chasing. 9:48
    • Internet problems, overheating and low memory cause the failures.
    • If the drop sits in low-end devices, ignore it.
  • Cached ads can expire before they are shown. 10:05
    • Preloading too many ads lowers the show rate, so reduce the frequency.
    • Call the is-ready function and move on to the next ad if it fails.
    • After a 30-minute session, a cached ad may no longer be valid, and Google won't count it.
    • The tip came from something Sid read recently.
  • Fast native refreshes lower the show rate. 11:16
    • Slow the refresh without hurting eCPM or revenue.
    • Developers tend to load and refresh native ads very quickly.
    • Run as many experiments as possible.

In the growth library

Sid Gupta, Co-founder and CEO at Appbroda, AdMob and AdX expert (hosted by Steve P. Young of App Masters). Explains the difference between AdMob match rate and show rate, what show rate to aim for, and a step-by-step process to diagnose and fix a low one. The benchmark bands (under 30, 30 to 50, above 50) tell you when it is worth the effort.

Lessons that draw on this:

More from App Masters · The whole library, and ask it anything

Transcript381 lines
0:00

what is up app Nation it is Steve p young founder of app masters.com the place you go when you want actionpack content related to helping you grow your app downloads and your revenues and today we're going to talk all about how to improve your ad mob show rate thereby increasing your admob revenue and joining me as always for anything related to admob and adx is Sid gupa he is the co-founder and CEO at APPA go check him out it is appa.com it welcome back thanks Steve how are you man I'm great my brother so I'm super interested in this show rates ad mob revenues super critical I mean show rates almost like the main component right it to increasing your revenues like if you can show more ads and fill those you know ad placements you're going to end up making more money so what have you guys discovered in how to increase your show rate cool that's a really interesting thought uh stepen I like i' like to add like one little

0:56

clarification there I think this is one really interesting fact that a lot of developers get confused about so there are two parameters that a developer should be worried about outside of the ecpm and the the ctrs these are show rate and match rate now the first thing to actually understand is the difference between each one of them let's take an example so you've got an ad placement let's it's a banner placement and there are 20 ad requests that you generate from that ad placement now these 20 ad requests go and hit Google Now of those 20 ad requests if Google says I have ads for 10 of them so 10 ads come back and 10 became unfilled so 10 out of 20 that becomes your match rate so match rate in essence is a total number of ads that Google has for all the ad requests that you have that's match rate now once you have those 10 ads let's say for whatever ABC reason only two get shown actually

1:47

in your app that means of the 10 total ads only two actually got displayed in your app so that's 2x 10 that's 20% show rate so just to simplify uh and repeat this again match rate is actually the total number of ads that Google has for your inventory showed it is a total number of ads that you actually showed in your app so there's a little bit of confusion a lot of time developers confuse the two terms and it's important to understand uh each one of them now what do you talk about match rate great so we we won't touch match rate match rate is a problem completely separate completely different to talk about today we're going to talk about show rate show rate and show rate primarily what I try to do today Steve I I thought I'll uh try and break it down into a couple of questions before the podcast I actually went on the internet I had to search for a lot of questions people keep asking and what I've done is I've actually

2:32

figured out four or five questions that people typically ask I've obviously reframed the language to make it more simpler and palpable for everybody so I thought I'll I'll talk about each one of them and try and address those concerns so the first question that naturally comes to mind should I be worried about show rate yes and no yes because it all depends on what is the show rate that you're getting now the next question to ask is what is a good show rate now unfortunately there is no standard a golden number that Google recommends it's a good show rate but my personal recommendation my personal learnings from multiple developers that I worked with I formed sort of a very Instinct Baseline that I understand now let's divide into three different categories and again this is all my personal recommendation so please don't quote me on it or Please Don't Ask Google about it but this is what I understand and can

3:21

say it's a fair Fairly good show rate or a mediocre show rate or a something to worry about show rate right let's assume your show rate is between 0 to 30 now that's something definitely you should worth look uh you should look into it is a low show rate and it affects it will affect your cpms and revenue to some extent definitely worth looking into you can increase them so as to increase some Revenue definitely if it's 30 to 50 it's mediocre it's not bad it's not good I would say bear it out there are a lot of things lot of experiments that you can run or spend time in improving your Revenue I would say put it at the low part in terms of your bucket when you have ample time when you have enough time and you're not working anything then look at the experiment run it for a long duration of time and get results but if it's above 50 don't even think about touching it it's it's not going to make much of a difference 50% is

4:10

considered really really good uh most cases I've seen is 50% above they're able to increment it but only in certain number of cases and the number of the chance of increasing it is really really low there are better ways that you can increase your Revenue so again should you be worried about sh it if it's 0 to 50 you should be if it's 0 to 30 definitely spend some time trying to improve it if it's 30 to 50 find some ideal time and you don't have anything in the pipeline and then try try work on it so let's talk about how do you figure this problem out how do you go about solving this problem and understanding what is the core reason for this what I've done is I've again uh built a step-by-step process to help you understand and sort of reach to a conclusion or an area where you can sort of drill down and find the problem so the first thing and I know it's the most obvious thing uh I'm just saying it

4:58

because a lot of smart folks would have already figured it out but for the folks who did not there are a couple of tools that Google has that allow you to check if your integration has been done correctly or not this is an add inspection tool and a mediation test tool these are really simple tools that allow you to add a test device figure out if the integration is correct or not I know it seems fairly obvious but sometimes when you are doing an update you might have goofed it up just check it out if it's good let's move on to the next step the second thing uh which you might have noticed Google has actually come up with a new report called an ads activity report which is a new report called the admob network report what they've done is actually given a lot of new parameters for you to play around with you can really do a deep dive into what's working and what's not working for the show rate try and deep dive

5:41

further into multiple parameters and see where exactly is the show rate going down right so you can look at uh the format is it a format problem you can look at the platform see if it's an Android or an iOS problem you can maybe do it in version number so you have multiple build releases see if there is a perter build which is leading to a lower show rate even within the formats you can further drill down by ad type that also is really really helpful in figuring out if maybe it's a sizing issue right and we'll talk a little more about that but a lot of times the ads that are being supplied back might not be being rendered for an image type or an ad type particularly right so do that uh and the final thing you can also figure out by ad Source a lot of times the other partners that you've integrated may not be able to render the ads for an ABC reason spend some time drill down on the ad activity report

6:29

also called the admob network report and figure out exactly the case where this exact drop in show rate is happening once you've done that the next thing I would recommend is something called error codes this is an old capability by Google which a lot of people use for different reasons but this is also an interesting way to sort of figure out what the cause of show rate drop is when you send an ad request to Google Google either responds with an ad or it actually responds with an error code there are about 15 16 error codes that you can look at and each error codes means something now you can actually add listeners that allow you to understand what is the error code that is being sent back based on that you can actually figure out if there is an issue with configuration or if it's an issue with ADD entry so these are two very interesting ways that you can sort it out once you figure out the error codes

7:16

uh once you figure out the ad activity reports the next thing is actually integrating Firebase I'm assuming you would have already integrated Firebase for a couple of other reasons but this is also an interesting way to actually solve the problem run a couple of ab tests see if those placements can be replaced if it's a version issue trying disabling the ads see if it works but Firebase is your answer for running a couple of ab tests and understanding how to finally solve that problem now guys I would recommend be extremely patient when solving this because this is going to be a long tedious process it's not something you can solve quite easy most people I've seen trying to solve this have taken a good amount of time to get to a final solution be patient find out a time when you have that bandwidth to solve this issue and then go at it having said this I also realize a lot of people ask me man can you at least give

8:02

us a couple of ideas on what could be the possible reasons what am I actually looking for so have I have some sense of what am I actually working towards so cool man I I I've actually built a list uh helping you understand the common issues that I have seen I have spoken to developers about right the first issue I know again it's quite obvious but a lot of times we have integration issues so get that out of the way like just check it out figure out if it's a version issue if a certain version is not performing well figure out what went wrong and then solve it another very Corner case in a lot of times when happens is you might have integrated a format and when you're actually there are certain ad sizes which are not being able to render for example there's a lack of space you've not you've integrated the banner or the native in a size that is actually limiting to certain devices that can be a problem so

8:48

one very interesting way to solve is you can actually upgrade a lot of your formats for example if you're looking at Banner and this particular issue is with Banner make it an Adaptive Banner uh these are new formats that Google allows where the width and the height are adaptable and that automatically adapts more sizes so Auto adaption works really really well uh one simple issue the second is placement issue a lot of times we actually Place ads on screens that do not come always for example if you built a game and you've got ads after every level but the game's really hard like most people crash after third level so if you've got ads on the fourth or fifth level which are actually loading but they're actually not being shown that can be an issue youve got to figure out maybe decrease difficulty level uh so that you can actually make sure the ads are visible the third is also a problem where a lot of times when the user moves

9:37

through the screen really really quickly there's not a lot of time and you've made an ad request but the ads not come back yet because of that the users move on to the next screen and the ad's not loaded yet and because of that the ad's not visible the fourth reason uh again something you can't do much about but there are a lot of time there are internet issues device overheating memory issues within the phone these are issues that I've also heard of can't do much about it usually a lot of lower end devices and this issue so when you're looking and segmenting your problems if it's lowend devices don't worry about it my recommendation another issue caching the issue that I mentioned uh a while ago was about if you've skipped an ad placement or if you've skipped an ad placement because the US a more fast through it right loading issue now a lot of time what people do is you do pre-loading ads you uh load a lot of ads

10:23

and show it keep the ads in the memory and show it when the user actually gets there you call the ad right when you do that you actually sometimes developers load a lot of ads so that can be an issue because uh your your show rate is less so reduce the frequency another interesting thing which I recently read about is that when you do caching and you get an ad now you do not know when you actually show the ad there are sometimes if your game plays really really long let's say the user is playing for 30 minutes and you already cach the ad when you actually show that cash ad at that time maybe the ad is no longer relevant right Google might have that ad is no longer a valid impression so if you show that ad Google will not count it as an impression Google has actually got a really neat function called is ready so make sure you load and call that function to ensure Dash tags are ready and are still visible if

11:12

not move on to the next tag call very simple problem right refreshing native and banner another very interesting problem we have a tendency to load and refresh native really really quickly so essentially it automatically affects your show rate reduce the show rate the refresh frequency in such a way that it does not affect your ecpm and revenue so that's another way way to do it again guys the key is running as many experiments as possible and be patient these are two really really key uh advice key things I will tell you to you find the solution and again understand the difference between matate and short it uh that's the takeaway I I would like you to uh leave it I love it that's my main takeaway said too hey should we give any how much time should we give for each test oh so it primarily depends on the number of AD requests that you're testing out so essentially as low as 10,000 ad requests you would figure out

11:58

what was working out better S anything else I missed you want to make sure we cover uh no man I think I've covered all of them once again it is apr.com if you're looking to increase your admob ATX revenues go check them out they have amazing case studies of apps they've helped increase their downloads Sid knows his stuff we've done a lot of content together so once again it is apr.com let them know you heard it on this very podcast so we continue to come back say thank you so much for coming on and doing this thanks for calling me Steve always a pleasure helping you figure out ideas on how to solve issues for IOP us you're my go-to guy and every time I have ad mob questions I always send people your way all right guys until next time I'll see you on the next video