My first campaigns ever

[matelo]

Hi guys!

first of all sorry for my terrible english! I just started running my first Cpa campaign ever using Facebook Ads.

I'd like to know if I'm on the right way and if each of the following steps is correct!

The campaign is structured like this:

- I picked 2 similar offers in gaming niche (payout: 2,6$ and 2,2$)

- Bidding: oCpm

- I created 8 adsets targeting 18-25 male from US who like games similar to the offers (4 adsets for each offer).

- I picked 10 images (the images are the same for all the adsets) and I created 2 different copies.

- I'm using one landing page

- I'm not using a tracking software



Results after 2 days:

Offer 1 (payout 2,6$)

* average CPC > about 0,10$ (the best adset got 0,08 and the worst 0,25)

* clicks on Facebook > 365 (expenses of the campaign> 40$)

* clicks on the CPA network > 130 (so the CTR of the landing is about 35%)

* conversions on the CPA network > 7 ( total earnings:18,20 $) The conversion rate is 5,4%


Problems> very low conversion rate


Offer 2 (payout 2,25$)


* average CPC > about 0,17$ ( I'm sure I can lower the cpc to about 0,12 by excluding the worst ads!)


* clicks on Facebook > 223 (expenses of the campaign> 39,24$ )

* clicks on the CPA network > 53 the CTR of the landing is about 23% (low CTR!)

* conversions on the CPA network > 5 ( total earnings: 11,25 $) >The conversion rate is 8%

Problems> Low landing ctr and low conversion rate



Should I stop the campaigns or is it to early?

Do you think they can be profitable?

Thank you in advance

 

[crysper]

Hey matelo,

Your campaign looks ok, these are usual numbers for a new campaigns, even though sometimes you may get very low performance.

So far you know that the offer converts(which is important) and the conversion rate is not bad(usually it's related with your targeting and ad message). This means that you can optimize the campaign and get it to green.

Landing page CTR is not that low, but you may get it higher.

Follow these steps:
1) Get a tracking software. You can start with Prosper which is free. Prosper202 Self-Hosted Affiliate Tracking Software & Apps You'll find details about installation on their website.

2) Once your tracking is set and working(since you are new it may take a while to install and test the tracker, but it's very important to have a good tracking in place), create 2 other landing pages, totally different from each other. Keep them simple: headline, image, 3-4 bullet points and big CTA(call to action). Later you'll see that we have on AffKit.com a lot of scripts to optimize the landers even more, but for now, start with 3 simple landers.

3) Cut the non performing ads( assuming that they have at least 5000-7000 views)

4) Try new ads similar with the best performing ones

5) When you have enough data (at least 20 conversions per lander) you can try our tool to know when to cut the landers AffKit.com > When to cut landers. Make sure you add only the landers that are for the same ad type, otherwise the results may get skewed.

 

[matelo]

Hi Crysper, thank you for your quick reply! If you don't mind I have a lot of questions


Landing pages

- At the moment I'm using a multistep landing page (like this one http://www.affiliatefix.com/threads/multistep-landing-page.17478/)

I have another simple landing (video, images, bullet point ecc..). As for the third landing what about using a "game review" landing?


- How to test the three landings?

Do I have to test them manually (eg. changing the landing after a few days or creating news adsets) or does the tracking software (I want to try Adsbrige) allow me to rotate and split test the landings?


Tracking Software

I'd like to try Adsbrige, it's pretty affordable. As far as you know does the free version (AdsBridge - Next generation tracker and landing page builder!) include lading pages split test?

Ads

my audiences are very small and targeted...isn't 5000 views too much?

how to test the ctr of every ad? Sometimes the best ad tends to prevail while the others ads never get viewed.

In addition to landing page should I also test each ad to understand which is more profitable?


thank you in advance!!

 

[crysper]

Both multi step and review landers are good as long you keep them simple. You have a very short attention span since the visitors came from a social platform.

I don't know AdsBridge, but I think they have lander split testing included in the free version so you don't need to split test manually.

5000 views is not much because you need to have statistical confidence that the ad is not working, otherwise you may cut the wrong ads. Many times(due to randomness or day parting) you may get many clicks on the first 2000 views and the next 2000 you'll get very few.

Yea, this is an issue with Fb, that some ads get many views and some very few, at least at the beginning. A solution is to create multiple campaigns with max 5 ads each.

 

[Rvfamily]

@matelo,

Thanks for putting together a nice informative thread. Keep it up.

As always,
Here's to your success

 

[matelo]

Thank you @crysper you've been very kind and useful!! Just the last question..

Beside testing ads Ctr, landing pages etc..should I track which single Facebook Ad the conversions come from to understand which ad performs better? or do you think conversion rate is not affected by the ads?

thank you and sorry again form my english

 

[Rvfamily]

Thank you @crysper you've been very kind and useful!! Just the last question..

Beside testing ads Ctr, landing pages etc..should I track which single Facebook Ad the conversions come from to understand which ad performs better? or do you think conversion rate is not affected by the ads?

thank you and sorry again form my english

@matelo,

I was just reading an article on Charles Ngo's blog that addresses this very thing (statistical significance). Hope you don't mind me jumping in and that you find value. Here is the link to the article.

As always,
Here's to your success!

 

[matelo]

@Rvfamily great!! thank you