Facebook, where is oCPM located?! (15)
Hi,
Where is oCPM located? Here?
http://prntscr.com/6z7qby
Or is it unique daily reach that is also called oCPM?
More questions I have:
Ok, although mobile is the shit, I am still curious about facebook. Anyways, I am doing some whitehat leadgen campaign in the netherlands. What I really need is a starting point. So testing stuff is something I wanna do later.
So far what I understood is (and please bare with me, it is just a general plan):
- If you are just starting out, and if you use a landing page, it is better to use 'clicks for websites', if you are having a working flow, it is better to go for post engagement.
- It is better to target as broad as possible, as long as it does not exceed the 4 million, the cheaper the CPM will be.
- It is wise to just copy your ad sets and spend 5-10 dollar max per ad set.
Right?
Anyone?
For now, I chose this one:
http://www.evernote.com/shard/s449/s...48b186374857e4
https://www.driftrock.com/blog/facebook-ads-bidding
I feel like that dog in the previous link haha
I still dont have a clear answer, it seems i dont see it when i select CPM -confusing-
Anyone?
bump
Anyone? or was this something that was in the past?
tralala
Can anybody help me please?
The impressions and clicks options are CPM and CPC bidding, respectively, as indicated in brackets.
The daily unique reach is completely different - ignore/avoid this.
However, when you pick "Post Engagement" in the optimize for dropdown pictured, it will say that you are charged per impression, and this is oCPM.
oCPM is CPM bidding which optimise toward an objective. Hence, the top option - which is an objective - is oCPM.
The particular configuration of oCPM in this case depends on your campaign's objective, which limits your choices for all ad sets/ads within that campaign.
Other things:
- If you are just starting out, and if you use a landing page, it is better to use 'clicks for websites', if you are having a working flow, it is better to go for post engagement.
I'm not sure of the rationale behind this - what you do after the click has nothing to do with the bidding mode. By working flow do you mean an optimised funnel? In either case, you want the highest quality audience for your campaign, people who give the lowest CPAs overall, or both.
This is not something that will always be X, or always be Y - it depends on the audience, the ads, what you are trying to achieve, your offer, etc.
It's kind of like saying that 320x50 ads are the best for Android offers, clearly not accurate.
For starting out though - if you are using a lander, pick Website conversions for your objective, then ideally have that conversion code from Facebook fire on your offer conversion (which should be doable on whitehat lead gen for sure!) or failing that have it fire when users click through on your lander. If you cant do any of that, just use website clicks as your objective.
You can test CPC or oCPM, they perform differently and should be treated as a split-test.
- It is better to target as broad as possible, as long as it does not exceed the 4 million, the cheaper the CPM will be.
Targeting broader often nets lower CPMs, yes, though this is then dictated by the demographic as well (this will fall apart if you say widen your age from 13-17 to 13-40 if the 18-40 bracket garners CPMs 3x higher than the 13-17 bracket... just by simple maths).
However, you will find that interest targeting (or similar) has a large influence on ad performance at the front and back-end. If your campaign can be tailored to people with specific interests, try to do that first.
There's no point making a campaign for home insurance and targeting all men 30-40 in California with an audience of 10 million, for the sake of broadness, when you could target all men in San Francisco, who are 30-40, have an income over 50,000 and are home owners and make a much more tailored campaign.
- It is wise to just copy your ad sets and spend 5-10 dollar max per ad set.
I disagree. This is one way to scale and its good for testing many iterations of ads but you'll find it tough to scale if you do so by giving Facebook hundreds of ads/ad sets with small amounts of data for each.
oCPM optimises at the ad set level and FB's systems also generally track/use the performance history of ads and ad sets.
If you split everything into small budget ad sets you will lose much of the optimisation that oCPM offers and get more volatile performance from the ads.
This is however something to test. Some people will find scaling with ad set duplication works better for them. Personally, and especially for whitehat leadgen with broad appeal to large audiences, oCPM with higher budget ad sets makes more sense.
Thanks Zeno! I am VERY grateful for your helping hand! Thanks!
Some questions still pop up in my head:
Originally Posted by zeno
Another question: when should I switch to CPC?
Originally Posted by zeno
Originally Posted by zeno
Originally Posted by zeno
Once again, thanks!
Hey there, many if these are really basic questions. I would suggest you spend some time training yourself up on Facebook first.
You can take this course https://facebook.exceedlms.com/student/catalog and then after you have passed all the courses, you will pretty much know the basics of how the Facebook platform works.
@ cmdeal, yes I will do that too, pinky swear. Still, if anyone can answer these last questions: that would be awesome!
My own experience in asking others for help has been that this is much more effective when the other party sees that I have done all that I can to try to figure out the issue myself before asking.
Just sayin.
Originally Posted by Chunk
1. Yes, absolutely. They are fundamentally different on Facebook.
2. You don't really switch to CPC at any point, it's just another bidding mode to test which has pros/cons, is more weighted on CTR, and makes you pay per action rather than impression (which results in you paying multiple times for the same user performing multiple actions). It's closest analogue in performance will be oCPM(clicks). Just another bid mode to test.
3. Yes. Only your data will answer the most precise questions.
4. It depends on the value attached to your audience. If you are selling insurance for Bentleys to a highly targeted audience of 5,000 then it may be worth it, even if you can only get 2,000 impressions a day, if this results in $100 spend and potentially $1000 in revenue.
But, as a guide, if you can reach them all in a day with under $100 spend then it's probably too small (so this depends on CPM for the audience!).
Home > Questions and Answers > General Questions