Karen Grace said
PS one more numercial question. Am I right in thinking that the cost per click on an ad is a mixture between the click through rate per impression and how many other people are going after that audience at any given time? Thanks.   

Correct. It's an auction. If you advertise in countries that few people advertise in, you will pay next to nothing per click, but no one will buy anything. The competition is the base factor, the ads performance is what gets you the traffic over the other guy, for more or less money.

Karen Grace said
Hi again.

So the good news is Bjork's audience is doing well- the best yet with the dynamic creative round. Gradually putting the budget up to get some solid data :).

So a question about pausing and un-pausing. Is it a bad idea to un-pause an ad that I paused a while ago? Should I clone it and start again? I want to leave the fb part as it is, but continue and experiment with squeeze page tweaks.

Great. Glad to hear it. Often pausing and unpausing hurts a campaign's performance. But not always. I always start by doing just turning it back on. If it doesn't do well, I clone it. Not much to lose.

On a similar note, how do you tend to test squeeze page tweaks in an ongoing way ? I was thinking of maybe letting a campaing roll and tweak something every 25 conversions or so and see which version of the squeeze page got me the cheapest batch of 25 conversions. Does 25 like a good amount to give me a good enough idea?

Thanks,

Karen  

That's basically what I do. For a more accurate split test you can use the split test script I included in an older split testing lesson in the IC, or use Facebook's proper split testing option (which is expensive). But I usually just do what you outlines. 25 is pretty low, and I would feel better with 100. But sometimes we just can't spend that much and I don't think it;s teh end of the world if we just decide early based on the best evidence we have.

Favour to ask... I realize you are breaking these questions up into individual threads per a request I made a while back, but what I meant to to request is that you break up unrelated questions into their own topics. Readers of the forum tend to browse based on topics and would never dig deep into a thread that didn't interest them. But each one of these questions is really it's own topic. So if possible, please break up unique questions into unique topics. It makes the info easier to discover and makes my work flow a bit easier as well. Thanks.

Awesome. Thanks.

To exclude subscribers I do two things... I export all of my subscribers from aweber and then import them. Then I create a custom audience of people who have landed on the subscriber thank you page in the last 180 days. Then I exclude both of those audiences from my ad. This way I just need to upload the new subscribers to the Facebook custom audience every 6 months or so to keep things up to date.

You don't need to do it on a campaign that is running but I would still do it on on your future campaigns. Facebook will know that your existing audience tends to engage and group them in your target audience otherwise. Or at least that seems to be the case.

To export...

If using legacy follow up and individual lists for customers vs non customers, I go to the list I want to export, click on subscribers, select "all subscribers", then export to a CSV. I then copy the email column and paste it in a text file without all the other data, and import that. But you don't actually need to do this. You can import the CSV directly. It's just one of those throw back things I still do for no real reason other than OCD.

If using campaigns (and a single list) you would go to the list in question, click on subscribers, then create a segment based on tags, ad tracking, or whatever you use to identify the segment you want to create a lookalike audience around. Then export, same as above.

To import...

Go to your ad manager and click "audiences". Then create an audience. Select "customer file", then import the file you exported from Aweber. To create a lookalike audience you then go back to the main audience page and simply create a lookalike audience (the lookalike audience is in a drop down below "create audience". 

All of that is from memory, so it may be slightly off, but it should be close enough to get you there. There are a few lessons on creating custom audiences here in the IC.

I do both because when you create an audience based on people who land on the page, you are limited to 180 days (365 in some instances). So that is why I do both. By uploading you get everyone that has been on your list for a while. By adding the landing page to the mix you get all the new people that you haven't had time to upload, making it such that you only need to bother uploading every few months or so to keep your audience pretty current. There also appears to be some inconsistencies in which people that land on the page don't overlap with your list, and visa versa. Not sure why that is. But by doubling up you get everyone (or closer to it).

Let me know if you still have any questions.