Has anyone tried to do manual optimization as opposed to the automatic optimization? From reviewing past data using the automated optimization, it looks like sometimes another network might pay more per click, but isn't the primary filler of ads.
Anyone have similar results or tips to maximize revenue?
__________________
--
Neque porro quisquam est qui dolorem ipsum quia dolor sit amet, consectetur, adipisci velit... Then why am I a programmer?
Has anyone tried to do manual optimization as opposed to the automatic optimization? From reviewing past data using the automated optimization, it looks like sometimes another network might pay more per click, but isn't the primary filler of ads.
Anyone have similar results or tips to maximize revenue?
I'm going to give it a shot. My mobclix eCPM the last few days has been really lame. I just reordered it based on my perceived performance over the last few days. We'll see how it goes.
I'm going to give it a shot. My mobclix eCPM the last few days has been really lame. I just reordered it based on my perceived performance over the last few days. We'll see how it goes.
i'd highly recommend against that. Sometimes I get some REALLY high paying campaigns (eCPM $10 or so) that comes through on a weekend. I'm not going to bother tracking that stuff and tweaking it manually.
i'd highly recommend against that. Sometimes I get some REALLY high paying campaigns (eCPM $10 or so) that comes through on a weekend. I'm not going to bother tracking that stuff and tweaking it manually.
If you're seeing higher earning with manual optimization, something's wrong with your set-up. The campaigns change too quickly even for us to optimize in time to take advantage. Burst campaigns can run for an hour. Manually optimized apps will miss this if you're not pointing at the correct network.
But as I'm telling everyone who will listen, the single biggest way to increase earnings 2-5X is auto-play ads...
If you're seeing higher earning with manual optimization, something's wrong with your set-up. The campaigns change too quickly even for us to optimize in time to take advantage. Burst campaigns can run for an hour. Manually optimized apps will miss this if you're not pointing at the correct network.
But as I'm telling everyone who will listen, the single biggest way to increase earnings 2-5X is auto-play ads...
3 questions.
1- In setup service tab, I have Allow-Auto-Play Ads set it yes for iPad 300X250. Is this right?
2- What is the best value for Ad Refresh Time in that page?
3- What is Auto-Play Interval and what's a good value for it?
To be honest, our eCPM seems to fall when we turn on automatic optimisation. A ton of impressions get delivered to MdotM which has the worst eCPM out of the networks, despite a CTR of close to 10%. (i'm talking about the 320x50 ads)
I'm not confident about how automatic optimisation actually works. We turned manual optimisation back on and eCPM rates improved.
Perhaps Adam can shed more light on the process, as he does encourage it.
As for full screen takeover ads, if you're going to active those, you might as well opt for integrating transpera, their video ads have a much better CPM rate of $20.
1- In setup service tab, I have Allow-Auto-Play Ads set it yes for iPad 300X250. Is this right?
2- What is the best value for Ad Refresh Time in that page?
3- What is Auto-Play Interval and what's a good value for it?
Thanks
Good questions:
1. yes.
2.& 3. 30-60 seconds but it doesn't matter because Auto-play ads are capped at once per 15-minute session to avoid user backlash.
on the auto play ads, can you control when they might occur, or do they just a child of the normal banner ad?
Yes, but if you'd like to control, I'd recommend adding another ad unit to your app (say 300x250) and enable autoplay on that unit, so when you show it you can expect auto-play.
We'll be replacing auto-play ads with full screen interstitial soon-- essentially giving it a separate unit.
To be honest, our eCPM seems to fall when we turn on automatic optimisation. A ton of impressions get delivered to MdotM which has the worst eCPM out of the networks, despite a CTR of close to 10%. (i'm talking about the 320x50 ads)
I'm not confident about how automatic optimisation actually works. We turned manual optimisation back on and eCPM rates improved.
Perhaps Adam can shed more light on the process, as he does encourage it.
As for full screen takeover ads, if you're going to active those, you might as well opt for integrating transpera, their video ads have a much better CPM rate of $20.
I'll forward on your comments about Transpera, it might be a good demand source for us to add to the exchange.
The optimizer is incredibly complex and even I don't fully understand the intricacies of how it works but basically it averages the earnings of your network feed over a period of time and gives that network priority for first call and rollover.
The only way it would earn less is if a network had a low average but would very occasionally send a very high paying impression. Our new RTB exchange eliminates this shortcoming by integrating demand sources that can bid real-time on the impression and we can optimize by impression, not ad feed.
However you shake it or slice it, you can't argue with results. If you're seeing better revenues with optimization off, then that's all the proof you need. For 99% of the people the optimizer will bring better results, but if you're the 1%, more power to you.
Just be sure to check back often to ensure you're changing as necessary...
I'll forward on your comments about Transpera, it might be a good demand source for us to add to the exchange.
The optimizer is incredibly complex and even I don't fully understand the intricacies of how it works but basically it averages the earnings of your network feed over a period of time and gives that network priority for first call and rollover.
The only way it would earn less is if a network had a low average but would very occasionally send a very high paying impression. Our new RTB exchange eliminates this shortcoming by integrating demand sources that can bid real-time on the impression and we can optimize by impression, not ad feed.
However you shake it or slice it, you can't argue with results. If you're seeing better revenues with optimization off, then that's all the proof you need. For 99% of the people the optimizer will bring better results, but if you're the 1%, more power to you.
Just be sure to check back often to ensure you're changing as necessary...
Adam, if you're taking requests, have you guys looked into Greystripe? Their ads look nice (actually 320x50), they don't take you out of the app, they seem to pay by impression and they pay reasonably well.
Adam, if you're taking requests, have you guys looked into Greystripe? Their ads look nice (actually 320x50), they don't take you out of the app, they seem to pay by impression and they pay reasonably well.
We're definitely in talks with Greystripe.
PS: Our SDK will have in-app video (without kicking you out) in our next SDK release...
Yes, but if you'd like to control, I'd recommend adding another ad unit to your app (say 300x250) and enable autoplay on that unit, so when you show it you can expect auto-play.
Adam, I won't be integrating the ads into my apps myself.. I work with a remote developer.
What exactly should I say when directing them on how to integrate the auto play unit (for iPhone and iPad apps alike)?
i.e. "when you integrate mobclix, please add a separate 300x250 on screen X and enable autoplay on that unit. all other screens should have 320x50 units that do not autoplay."
Firstly I would like to thank you for being so responsive regarding these concerns. While Auto Play ads can be quite beneficial for developers, they can also serve as a hindrance. I placed Auto Play ads in my application for one day, and received more 1 star ratings than I did in 6 months of just displaying normal banner ads. Some apps are not made to host full screen interstitials, and customers had no problem letting us know that. It comes off a bit invasive at times, and it locks the customer into watching something that they may not want to see. What strategies would you recommend for those that are unable to host Auto Play ads?
Firstly I would like to thank you for being so responsive regarding these concerns. While Auto Play ads can be quite beneficial for developers, they can also serve as a hindrance. I placed Auto Play ads in my application for one day, and received more 1 star ratings than I did in 6 months of just displaying normal banner ads. Some apps are not made to host full screen interstitials, and customers had no problem letting us know that. It comes off a bit invasive at times, and it locks the customer into watching something that they may not want to see. What strategies would you recommend for those that are unable to host Auto Play ads?
There's a reason autoplay pays so well, it's a rich user experience that demand the user's attention. It's not something you can show as much as a banner. Thusly, it important that you-- the developer-- manage the user's experience. Placing too many autoplay ads is worse than placing too many banner ads. It's invasive and annoying.
The strategies that I see work best are placing the ad during a time when the user is finished with something (a level for instance). Keep in mind the current version of non-iAd in-app video is a separate application, so it starts the media player and actually removes the user from the application. Not a problem if you've saved progress, the user can switch back, but you can imagine how awful that would be to use multiple times.
Our next SDK (due out in two weeks) will have a new ad-unit: an in-screen full page unit that will take the place of the auto-play ad. This will allow you to place an ad during a break and not force a video onto the user (it'll also be in-app too, no more having to switch to the media player), so keep that in mind for how you'd like to use it in your apps.
So long story short, place the ad where YOU wouldn't mind seeing it in your app. Don't let it take over the app-- in fact I recommend you use it as a reminder that the app is ad-supported and the paid version is ad free.
Oh, another thing, the 320x50 unit will only autoplay once per session, the300x250 will autoplay ad many times as you show it (and there is demand available).
Adam, I won't be integrating the ads into my apps myself.. I work with a remote developer.
What exactly should I say when directing them on how to integrate the auto play unit (for iPhone and iPad apps alike)?
i.e. "when you integrate mobclix, please add a separate 300x250 on screen X and enable autoplay on that unit. all other screens should have 320x50 units that do not autoplay."
Will they 'get it' if I word it that way?
Thanks!
Good news here: it's a "yes" button in the dashboard. There's nothing that needs to be done in the app to make it work (provided you have SDK 4.1+ of ours).
Services>Set-up>"Allow Auto-play ads" turned to yes!
This is assuming you already have the ad units enabled and placed (as discussed above) in a place that is acceptable for auto-play to run.
I have always found that automatic optimisation is better..to do manual prioritization you need to have good understanding of the market & networks also it needs constant monitoring….
I am finding that some of my apps get a high percentage of custom ads served even though I only have it set to 1% in optimization and at the bottom of my priority list. Optimization is currently set to automatic.
The app I am talking about has a CTR ~3% but for a 728x90 iPad ad I get 25%-35% of my impression served up as custom ads. I also have 6 or 7 ad providers active. Has anyone else experienced this?
I am finding that some of my apps get a high percentage of custom ads served even though I only have it set to 1% in optimization and at the bottom of my priority list. Optimization is currently set to automatic.
The app I am talking about has a CTR ~3% but for a 728x90 iPad ad I get 25%-35% of my impression served up as custom ads. I also have 6 or 7 ad providers active. Has anyone else experienced this?
This is most likely because you aren't getting high fill rates and the custom ads are picking up the slack. There's really no reason you shouldn't have all networks for a particular ad-unit turned on, the optimizer will call the highest paying networks first and iterate to the lower paying networks. Naturally, custom ads (that pay nothing) are called last with automatic optimization on. But you can change the allocation of how much traffic are directed to custom ads.
I have always found that automatic optimisation is better..to do manual prioritization you need to have good understanding of the market & networks also it needs constant monitoring….
There's a reason autoplay pays so well, it's a rich user experience that demand the user's attention. It's not something you can show as much as a banner. Thusly, it important that you-- the developer-- manage the user's experience. Placing too many autoplay ads is worse than placing too many banner ads. It's invasive and annoying.
The strategies that I see work best are placing the ad during a time when the user is finished with something (a level for instance). Keep in mind the current version of non-iAd in-app video is a separate application, so it starts the media player and actually removes the user from the application. Not a problem if you've saved progress, the user can switch back, but you can imagine how awful that would be to use multiple times.
Our next SDK (due out in two weeks) will have a new ad-unit: an in-screen full page unit that will take the place of the auto-play ad. This will allow you to place an ad during a break and not force a video onto the user (it'll also be in-app too, no more having to switch to the media player), so keep that in mind for how you'd like to use it in your apps.
So long story short, place the ad where YOU wouldn't mind seeing it in your app. Don't let it take over the app-- in fact I recommend you use it as a reminder that the app is ad-supported and the paid version is ad free.
Oh, another thing, the 320x50 unit will only autoplay once per session, the300x250 will autoplay ad many times as you show it (and there is demand available).
It's a much more popular way of giving a rich-media experience to the user. Non-interrupted. Since the fill is relatively low from full screen (new and expensive unit) you'll see a less-than-ideal fill rate. So we automatically back-fill with 300x250 when a full screen isn't available.
On the full screen you're seeing a 3-5X increase in CPM from a 300x50 and a 300x250 is generally 1.5-3X CPM.
mobclix optimization
Linear Mode
