FACEBOOK ADS: Conversion Goal for higher up the funnel (8)
Hi Friends,
was reading vortex's fb guide
''So my suggestion would be to just ignore the learning phase for the most part.
YMMV, so again, DYOT. If you're wanting to try to exit the learning phase, but want to do so at a smaller budget than what I've illustrate above, consider optimizing for events further up the funnel than the final conversion (e.g. Purchases).
For example, for ecom, try to optimize for IC or ATC. For affiliate offers, optimize for landing page CTA clicks to the offer. If you're REALLY strapped for cash, you can even try to optimize for video views. The cost associated with these actions are less than the cost required for purchases or offer conversions - which will require a smaller budget to get the 50 conversions a week to exit the learning phase.''
Would like to seek you guy's opinion.
There seems to be two schools of thought for ecom
One advocates only using purchase as conversion and never using other goals
One advocates using goals higher up the funnel and train the pixel so video views > lead > ATC > IC > Purchase
appreciate any advice on which is actually better since ios14 ?
so as your idea, we need to optimize step by step from views > lead > ATC > IC > Purchase, and after 50 conversions in each stage, we move to the next?
in the other hand, my friend usually just train the pixel learn the PPE phase and then go straight to purchase phase
it seems like optimising for IC can sometimes attain better ROI than purchase.
how is it working for your friend when he trains with PPE first?
I'd love to know what the community thinks as well!
Just to clarify though: I would advise to stick with the Conversion Objective if conversions are what you're ultimately after, so FB would target audiences that would be the most likely to give you conversions - even if it means you won't be making enough conversions to exit the learning phase.
BUT - if you're adamantly wanting to exit the learning phase - AND you don't have enough budget to get the required number of conversions - then optimizing further up the funnel would be the only way.
Maybe I should take out that suggestion entirely, just to avoid confusion.
Amy
Originally Posted by vortex
how do i know if im not converting on a positive ROI because of the pixel being too new and too little data vs the product not converting?
It's very strange I could use the very same adcopy and creative and run ads and there would be no sales,
but supposedly this same adcopy and creative in this example is run by others and converting at a positive ROI.
How much purchase data is sufficient for the pixel to convert properly assuming the product is a working product?
pls forgive me if my question is too noob as i am still learning haha
Originally Posted by schlossy
I simply set up my tests in a way that I'm testing multiple angles and ads for each audience, using multiple adsets to hit multiple "pools" for that audience.
And I'd give my best shot at coming up with audiences, ad angles and ads for this initial testing. This includes adequate research into the niche, product/offer, angles, audiences, everything.
To me, by hitting multiple audience pools for each audience, I'm "making up" for the pixel being new. In other words, instead of waiting for FB to gradually test and refine the target audience, I'm "forcing" the algo to sample multiple audience pools right off the bat.
Then - if in spite of testing multiple audience pools using multiple ad angles and ads, for multiple audiences, I don't see promising stats very quickly, I would usually just test another product.
Yes - by using this approach I MAY miss out on products that could turn out profitable if I had spent more budget and time on priming the pixel etc., but I'd also be saving money on potential dud products. There are pros and cons and you just got to find your balance between accuracy and efficiency.
Having said that: Intel is king. If you have an affiliate manager or another affiliate that you trust, that tells you a product or offer is converting like gangbusters, then you may want to test more ads and audiences and run traffic for longer to give the campaigns more of a chance to perform.
The disclaimer here is that I haven't run any split-tests to see whether it would be better to a)set up multiple adsets to test audience pools, or b)run a single adset for longer periods of time to let the algo and pixel do their thing. I suspect that such split-tests would have different results every time anyways.
Another important thing to consider too is how confident you are in your ability to create ads for that particular product/offer. If you're operating in a completely new niche, you may just need to burn money for a while by picking an offer and just test everything you could think of and over-run some stuff to make sure the lack of conversions is because the product/offer+audience combination is truly bad. On the other hand, if you have enough experience with a niche that you already know what the best audiences and angles are, simply putting your best ads in an adset and duplicating it a few times to test different audience pools should be enough to tell you whether a product/offer is good in a short time.
Yet another crucial factor to keep in mind is the SIZE of the audience you're targeting. If your audience size is small and targeted to start with, then there's not a heck of a lot of "priming" that can be done for the pixel. If you don't get conversions right off the bat, chances are running it longer won't result in significant improvement. Whereas if you're targeting a broad audience and letting the FB algo zoom in on the best-converting pool(s), then it could take the pixel some time (and data) to sample different pools.
Also keep scaling potential in mind. If the pixel can only identify a profitable audience after a lot of sampling and a lot of data, it may mean that your creatives and/or product/offer are only appealing to a small subset of the audience being targeted. This is why it's so important to do research on audiences and angles. You need to make sure you're displaying the right ads to the right audiences.
Another disclaimer is that the FB algo is changing ALL the time. This is what makes it difficult to make suggestions. I would suggest to consider the above factors and just test different approaches. Yes you may "overrun" some campaigns the first few times by spending more money than you need to, but it's a learning process.
Amy
Start with conversion events that you are targeting, such as purchase or lead. Then check you front end stats to see if the ad is performing, CTR, CPM, CPC, freq. If the front end stats look good, but you are not seeing enough conversion events, then target something earlier in the funnel, such as ATC or IC.
We also focus on conversions. You know conversions are so valuable that they can make or break a business.
Home > Paid Traffic Sources > Facebook & Instagram