How To Analyze Facebook Ads To Get The Lowest Cost Conversions

Here’s the campaign I made yesterday.

Recently created Facebook ad

After one day, it’s got 120 conversions, spent 177 dollars, average of $1.48/conversion. What’s exciting is there’s a lot of good opportunities in this campaign to do better. Let’s take a look at how to optimize campaigns that are running for the best conversions.

You can see the global Udemy is giving me 37 cents per conversion, that’s sweet. That’s much lower then the cost anywhere else and it’s quite related to what I’m doing.

The first step I will do is go through here and turn off everything that’s horrendously high. I’m calling that over $2, that’s well above what I want. If it has no conversions – off. It costs $4.50 – off. Free software global – off. Global consultant looks good at 64 cents – for now, keep that on. Global learning – off. Global shopping – off. Hobbies and activities – off. Global Tech – off. Global Business – off. US Fiverr – off. US freelancer, I’ll keep it on pending a further report.

The second step is to try and analyze and see if any of these other campaign ad sets are doing good and just not driving a lot of conversions which is possible.

US Digital Marketing – off. US Zip Codes – off. US Coupons that’s doing very good too. US Discounting – not doing so good – off. US Zip Code where Udemy’s located – off. US Palo Alto – off. Eckhart Tolle – off. Thanksgiving – off. Freeware – off. Labor Day – off. Training and Publications – off. US Udemy isn’t doing nearly as good as global, but worth keeping for now. Hobbies and Activities – off. Tech – off.  Learning – on.

Now notice that Gratitude actually did work. Not something you might have or I might have figured. E-learning at the same time, did nothing – that’s off. US Student – off. US Business and Tech – off. Now, let’s refresh this screen and see how many campaigns I’ve turned off here.

24 out of 38, I’ve already cut 2/3 of the most wasteful campaigns in here and now what I have left are less than $2 per conversion. With the average currently at $1.48, just by continuing to run this their likely to drop.

What I really need to know is what else is happening besides conversions? It’s possible I’m getting good things to happen besides conversions. I’m going to zoom out here so I can see this more clearly.

Let’s take a look at the details, I’m actually getting 22 cents per click which is pretty good considering these are all sales countries and most of them are reaching great audiences.

Let’s see who’s getting the worst costs per click because those need to be paused. That’s a really atrocious one right there.US Tech Desktop seems to be singlehandedly causing the problem in the US tech campaign because the US tech mobile campaign was getting low cost clicks yesterday. That’s the worst offender right there – US Desktop version of the tech campaign.

I will open each one of these that’s the worst offender that I haven’t paused already. All of these other ones have been paused, except the entrepreneurship one’s not paused yet.

Let’s take a look at each of these so you can see that one click on the desktop was devastatingly expensive, but I’m getting lots of clicks on this. What actually am I getting out of it?

I’m getting a lot of website clicks according to Facebook but very few of them are truly loading the page. I only got 9 loaded key page views out of that. At a $1.44/conversion that’s not really good.

Let’s check the detail reporting on this ad set:

analyze Facebook Ads

That’s what it shows me but I can also look at and see okay, what else happened? I can take a look at page engagement, post engagement, post likes, comments, shares and see if anything else happened besides what happened on here.

analyzing the other factors that effect good campaigns

Since I’m getting a lot of junk clicks in that category, I don’t know if I want to keep that doing. That includes yesterday and today. One thing I can do is just do today and see what happens.

Same thing today, I’m getting a lot of junk clicks in this one. There’s 62 clicks that aren’t really loading much when they go to the page. They’re clicking the website, but they’re not actually even loading the page. To me, that’s a waste.

I’m very critically scrutinizing the conversions over a dollar. This one I’m going to pause because ultimately a dollar for someone to go to my website is not sustainable and all I care about are sustainable.

This US Entrepreneurship one had an interesting issue here. The amount of desktop clicks is really expensive as you can see over here. Newsfeed on Desktop the cost for 1000 was ten dollars and the cost per click was 1.49, but I didn’t get a lot of impressions on the desktop.

Let’s scroll down here and see where my lowest clicks are because that’s what I really want to know. Where am I getting the cheapest clicks at? It looks most consistently I’m getting external ads on mobile devices.

One thing I can do to clean this up is take out some of these columns I don’t need. I don’t need unique clicks, I don’t need unique click through, I’ll take off 1000 people reached, take off cost per unique click, I’ll leave frequency on for now, but I don’t need both reach and impressions and frequency, which is 3 I know.

I hit apply and now I’ve got a more concise report that’s easier to understand. The frequency is still pretty low on most of these. What I want to see is are any of these campaigns are driving things like Page likes or something that I might not suspect so what I really need here also are conversions. Ultimately cost per conversion, so I can put that back on here.

Since none of the frequency are too high, I’m going to take that off. For these reports it’s very important to make your ad interface streamlined here so you can easily see what you’re trying to look at because you could see before when I had all those numbers, it gets to be a real pain and if I just throw the campaign name down there at the bottom

Facebook seems to want a campaign name on there because they keep putting it on here. I just leave the campaign name on the end there because I don’t really need that, these are all the same campaign. Now I switch to per conversion and what I want to see are there any outliers in the data? Are there any campaigns that are working really well except for getting conversions?

The global free software was getting a lot of clicks and that looks like it was actually getting page likes, which I don’t really need as much page likes or post likes. This got 10 post likes, I don’t really need those. It’s already got 26,000 likes on the post. What I need is Facebook to find people to go to my website.

If I got 53 clicks on it even though it only works out to 13 cents/click and that’s a good price, but I want people going to my website out of it. So I paused that campaign because it was not sending people to my website.

I keep looking through over here and see are there any outliers that happen to work well? Some of these out of the box categories I tried worked real good for cost per click. You can see this one’s only 9 cents per click right here.

I’m scrolling down now. What I ultimately want are campaigns that bring lots of clicks and conversions. These 9 cents per click, I got good click rate, but almost no one actually went to the website so if they’re clicking on the ads, these are junk clicks then for the most part. Let’s take a look in more detail and see what Facebook says happened out of this. I go through here and  there almost all website clicks.

I got theoretical website clicks, but I didn’t get actual page views. Here’s the campaign that’s pretty sweet. US Tech external ads on mobile devices. The interesting part is I’m getting tons of clicks for a low costs, but again almost no real page loads. Those are wasted junk clicks.

Sometimes theoretically they could hop on the page and leave so fast the code won’t load. If that’s what’s happening, why would I want those clicks? Junk clicks right here. The US Tech campaign is filled with junk clicks.

What I want to know is anything else happening besides people going on my website. You can see it’s almost all just clicks to website with a really low conversion percentage. That’s why I paused that campaign, don’t want that. I paused that ad set. The campaigns used to be ad sets but now they split them and ad sets are much more helpful for getting good results.

Here’s an interesting one – global free software. I’m getting 2 cent clicks on these. Hardly anyone’s actually loading the page though so these external ads seem to be working really well aside from the fact they don’t actually load anything.

Again, getting post likes, but I’ve already got 26,000 post likes, don’t really need that. That campaign is paused. A lot of cheap clicks are available, but the quality on these clicks seems to be terrible.

Scrolling down now we’re getting into the really good opportunities here. If I’m able to get clicks for 6 cents and people are genuinely loading the page, and they’re external ads on mobile devices, it doesn’t matter to me where they came from if they’re actually loading the page though.

That’s what’s working on the coupons. I’m getting a ton of clicks to it but a lot of people actually loaded the page on it. That’s where I’ve got my cheapest conversion down here consistently is in the coupons, but on the global consultant right hand column, I actually got 2 clicks for 3 cents somehow and I got one conversion. One person actually clicked on it, they did two things they liked the post and then they went to the website. That’s pretty cool, that was a very low cost, too.

The issue that I’m having is that most of the conversions are happening on mobile and do I even really want mobile conversions and page loads? Are people even going to take my course on mobile? That’s a good question.

Yesterday, it seems they are enrolling in the course on mobile, but the thing is to keep running the test. I want to turn this into a sales campaign so when I make it a sales campaign, then I’ll be making actual money out of it and it seems that if they already have the Udemy app  installed. They’re liable to make the purchase in it just as likely as if they were on the desktop.

Then again, it’s all theory. I don’t know, I have to test all these things out. What I do know is, I want the lowest conversions on the page then because if no one’s going to load the page, then it’s not useful.

The campaign here seems to have already worked good. I just want to check on these dollar per conversion and see if I can figure out exactly what’s the best way  to go about it is. If I go to age and gender, I might be able to get some additional insights here. I look over here and I try and figure out exactly where did I get the lowest cost conversions by age and gender?

Then I might be able to target my ads better. I’ve got unknown genders here and I’ve got 35. The age isn’t seeming to make a huge difference, it seems the placement’s making a bigger difference because I’ve got male, I’ve got female, I’ve got different age ranges all across here. It doesn’t seem to make a big difference age and gender.

Let’s try country. On the country I’ve got one conversion, that’s the cheap one from Singapore, a couple of cheap one’s in New Zealand on Udemy. That’ interesting.

New Zealand tends to be good because of its small population but there tends to be a lot of excellent wealth opportunities there in terms of the kind of people that are there. I might just want a New Zealand only ad set based on this. I’ve only spent 6 cents in New Zealand and I’ve got two conversions out of it on the Global Udemy. That’s pretty good.

Now I keep scrolling up and see where have I got conversion that scale? The US coupons is where I’ve got the coupons at the best scale. And the US Entrepreneurship after that. Let’s see if I kept that actual campaign on.

If I’m doing well on something, I want to make sure not to turn things off. I’ve got some good conversions for under a dollar here. Now I scroll up and find out what am I doing terrible? Mostly it’s n the US Campaigns but also on Katar, all of these are doing poorly up here with expensive conversions but they are at least converting.

What I can’t see in these actual conversions is all of the countries here that are not converted. What I want to do now is sort by amount spent and see where are the countries that are the worst offenders in not converting at all.

US has several campaigns with no conversions at all, and then it looks like UAE I’ve got some conversions and no conversions on others. I keep scrolling down and if I’ve only spent 50 cents there it’s hard to even say whether I’ve got anything worthwhile done there or now. It looks like the US campaigns are the worst offenders for getting nothing.

Finally, I want to see clicks. Which have I got the most clicks out of? It’s the US Tech campaign. That’s interesting in a sense that I want to actually leave that running a little bit longer if I can.

The problem is the clicks are twenty cents each though and I want to make sure all these expensive clicks have been paused and I believe I paused all of them except the global Udemy and that’s just in Great Britain that it’s very expensive.

Now I want to go back to placement and do the same thing. Look at my cost per click by placement:

how to arrange your facebook ads when you're analyzing them

See if there are any low cost clicks I’m getting on the desktop. It seems I’m getting very few low cost clicks on the desktop. Almost all the ads are in external ads.

It looks like the desktop ads are fantastically expensive right now. Likely because of the time of year it is. Almost all the ad clicks I’m getting are mobile and if I can get a decent mobile click that converts, that’s worthwhile right now.

This is how I analyze the campaigns. I run through here and I ask myself: What is working and especially what is working good for driving conversions? The US coupon mobile ads on mobile devices. I literally might want to copy that verbatim and make a campaign like that.

In the next lecture what I will do is try and expand on some of these while continuing now at the end of this to pause some that are not working good.

The US Freelancer I haven’t seen anything noteworthy on that. Us Online Advertising – nothing noteworthy on that.

Odesk and the US learning a little over a dollar again, I don’t even remember seeing that stand out anywhere in the actual campaigns. I didn’t notice it had a low cost per click or anything like that. I can check and see what the cost per click on it is.

At $1.18, even though the conversions are high, those clicks are miserably expensive. Now are there any clicks on another area? US e-learning, It doesn’t look like that’s worth keeping. I’ll go through here and pause US learning.

I’ll actually look to make a better US Tech Campaign because the US Tech campaign is interesting because the desktop clicks are $2 each and I actually spent $5 on that so that’s awful, but I’m getting a lot of clicks on the US Tech Campaigns in other areas: newsfeed, mobile, on desktop, those were 51 cents per click – again awful, but the US Tech Campaign on mobile devices with external ads, that was going really well in terms of cost per click.

It’s interesting to see all of those different dynamics in one actual campaign. These clicks are only 8 cents each so what I want to do is pause this campaign and try and make another one with just external ads.

I’ll make a quick note pad for my next campaign and this way I can then remember which things I want to try to see if I can improve. I can see what is working the absolute best and go through here and just try and make another $5 a day campaign on those alone.

When I go to cost per website conversion, I can scroll down and then expand my notes on these to put the absolute things working the best and try them in their new ad sets so that’s what I’m going to do next.

I appreciate you watching this. I hope this has been helpful. This is a lecture in response to people saying “how do you analyze Facebook ads?” The analysis is where you make these big improvements over time.