Home > Paid Traffic Sources > Facebook & Instagram

---

A CPI Gaming Offer from Avazu/KR/$2.5 - First FB Campaign After Our Pivot From Pop (36)

---

Facebook is so different, after spending over a week reading threads from this board, I have decided to do it the good-old-fashioned way of learning - the follow-along!

I have just set the campaign to launch at 12AM - Vancouver time where I am located.

Campaign Structure:

Budget: $30 lifetime budget for each ad set, $90 total for the three-day campaign

Audience (potential reach 390,000 people) -
Demographic: Men 18-25 in South Korea,
Interest: mobile games + role-playing games (must match both interests)

3 ad sets targeting the same audience, with same copy and budget but different creatives (2 images, 1 video)

Each ad is duplicated twice within the ad set following the common practice

Ad set 1(image)- AD1, AD1-copy, AD1-copy
Ad set 2(image) - AD2, AD2-copy, AD2-copy
Ad set 3 (video)- AD3, AD3-copy, AD3-copy

Bidding: automatic bidding CPC

This is my first FB campaign, so the purposes of the campaign are to

1) learn FB marketing
2) test the creatives
3) build a profitable campaign if things go well

Therefore, I might pause the campaign early if I spot a winner, or not depends on what advice I can get from you guys.

Will upload a screenshot once there is data worth looking at.

---

You can definitely learn facebook ads from this way, but $2.5 for KR is not gonna work.
The best way to run app install campaign is app install ads which you cannot do it with a aff network.

---

First of all, thank you for the comment

Can you tell me why $2.5 for KR is not gonna work, do you mean the payout is too low, it will never be able to support the cost of the traffic?

I am aware that the best way is not with a aff network, but we need to start somewhere and we have to work with what we've got, right?

---

I have got the first data that is worth looking at, please take a look at the screenshots as follow:

Ad1


Ad2


Ad3



No conversions. And the relevance score seems horrible.

CPC seems OK-ish.

What should I do now? Keep them running? Would FB algorithm works its magic and find better audience for these ads?

---

How much do you get paid per install? For a free app/game, that CPC looks quite big and a relevance score so bad makes me think this campaign is doomed.

You have to up your relevance score/find a better audience where your offer is relevant. The super low relevance score is also often a sign of bad creatives. People don't get it, don't like the creative, etc. You should also check if people hide your post, or even worse, report it as spam.

---

The payout is $2.5.

Please take a look at the stats of all ad posts, they look fine:





I'll take it as a sign of bad audience targeting then. So I targeted Men in Korea, 18-25 who likes mobile games + role-playing games, is this bad? And should I pause the campaign now?

---

It could be a combination of bad ads and bad targeting.

Interest "mobile games" and "rpg games" is kinda tricky. This audience may contain a chunk (or a big percentage even?) of advanced gamers which have no interest probably since they are all up to date on the latest games, thereby will not fall for an ad like this I think. Perhaps they are not even interested in these kind of mobile games, but going for A-level games only (perhaps this could be an interesting angle to approach them).

I know in the Netherlands a lot of bored housewives are playing games a lot, not sure how that is in Korea ofcourse so don't pinpoint on that. Ask advertiser/network what are the best audiences maybe? If thats really men 18-25 I suggest to replace interest with similar kinda games, which probably attract similar audiences.

But if you think the targeting is on point, you must definitely optimize your ads if you want to continue with this offer particularly.

---

I was expecting you

Solid advice! I will take them. Since I basically copy/paste the official creatives of the game itself, they can be bad, but they can't be that bad. Therefore, the more logical explanation would be what you described here - bad targeting. I'll still let the campaign drain its budget, because I want to see how much the FB algorithm can improve the relevance score. Would this be a bad idea?

---

Well, patience is one of the key things in affiliate marketing. Honestly I never gave a relevance score 2 a slight chance to optimize. I doubt it will go to 7+ eventually.
In general I like your attitude, besides hoping it would go better after a while, you still gather data and learnings from it somehow.

---

Hey, I take that as a No. I will either
1) pause the campaign
2) reset the targeting on the fly
3) create a new campaign using the same creatives, but different audience targeting

Which one would you do?

---

Then I suggest to do #1 and #3.

---

Generally sharing your creatives too will help getting more productive input - you should test plenty of them but some experienced marketers can tell when creatives don't fit the success framework AT ALL.

It's hard to say whether you should change audience or creatives, or perhaps both!

---

Here they are




And this is the video I ripped from YouTube
https://www.youtube.com/watch?v=a1oAMuJHSA4

Please advise

---

Definitely test more images, in-game screenshots, person playing the game, images which show more action in general etc etc

---

So I should adjust both the targeting and creatives then? Ok, I'll pause the campaign and re-launch with new targeting and images.

---

It's still unclear why you have a bad relevance score. Changing both at the same time would make it hard(er) to see where the trouble begins.
So best to choose either one of them first.

---

Makes sense, I'll make new Ads using different images to see what happens.

---

I just checked my Facebook newsfeed, saw about 5+ game ads (yes, I'm a hardcore gamer), all of them use in-game video clips, I think I'll make a couple of them just like those and test again.

---

On top of all the reasons described above, this is an old game, the potential might be limited as the market gets saturated. I just received a recommendation from the manager of a new game which looks amazing (I would play it myself), so I'll launch a new campaign testing the new offer using what I've learned here. I might or might not start a new follow-along there, but this has been awesome, thank you, guys!

---

You should continue this one, since it's a similar offer and we now know what you tried before!

---

Ok, I'll take your advice and will re-launch it with new creatives this time.

Actually, I was just told by the manager that this offer no longer accepts Facebook traffic. But I will continue the follow-along using the new campaign I just launched, I've just got a very similar result as the old one, very similar situation where I got a RS of 1, and need to find better targeting/creatives. Here is the stats,

AD1


AD2 hasn't yet to reach 500 impressions. Therefore, there is no RS to look at, but I would imagine it to be similar.

---

After re-visiting this post https://stmforum.com/forum/showthrea...eat-sheet-here of zeno's

I think more than likely it is my targeting in the wrong here, so I'll do more study/research to come up with better targeting first. As a gamer myself, I think although there is room to improve, the creatives are not the bottleneck here.

---

Be very very careful saying this. Really, take it more serious.

---

You are right. But I was just referring to the current situation, it is more likely the targeting is in the wrong, because my knowledge of it is next to zero, I should focus more on it instead of the creatives now. One variable at a time principle. Once I get a more solid understanding of targeting, and the RS is still horrible like this, I'll test as many creatives as I can.

---

Just saw this post by caurman https://stmforum.com/forum/showthread.php?24204-How-To

I'll use the tricks to make more images and test. After all the readings, it now comes the most fun part, test, test, and test!

---

Congrats!!

Here are my 2 cents:

1. Constantly check the integrity of the offer. CPI offers get shared by the aff network to other affiliates constantly so you might run out of budget and not know it.

2. Consider running a parallel CPM campaign.

3. You can try and increase the PO by finding CPA offers for apps and testing them out. Registration, Completion of tutorial etc.

4. You said it but i can emphasize it; Old titles versus new titles. Tons of old apps out there that are saturated.

Good luck

---

Solid advice, but I can't seem to understand the point 3. What is PO? And what do you mean 'Registration, Completion of tutorial "?

---

I got 2 conversions after re-targeting using a new approach, the ROI for that ad is around -78%, I think it has the potential to profit after optimization with landers. However, the RS is still very low, only 2.

It appears to me the targeting is Ok-ish since there are 2 conversions, so the most of the problem would be among creatives. Which means I should focus more on finding/making good creatives and test more.

Can I assume that, once I manage to boost the RS, the CPC will drop accordingly, making it much closer to positive ROI?

This seems promising if yes.

Thank you again, guys!

---

PO = Payout

Some advertisers are creating offers that fire the pixel not upon the install of the app but rather when the user completes a different task such as:

1. Registration of email
2. Creating a charachter
3. Completion of the Tutorial level
4. Reaching Level X

These result in much higher payouts. Now because you are supposedly bringing higher quality of users its worth testing some of these offers...

---

Although not understanding it 100%, I can smell that these are coming from real world experience, I will keep them in mind!

---

Yes, thats correct!

---

Nice thread.

While my FB experience is still developing, I have tested a lot of different methods. Some thoughts from my experiences that you may want to consider.

The biggest challenge in the beginning of the campaign is narrowing down to the right audience both with your targeting and your creatives/messaging.

1. Video drive lots of attention and engagement > retarget view views at 25%, 50%, 75%
2. Another good thing is to try are the slideshow options. Load in a bunch of different images and it views somewhat of a video interactive element (but its not video) and you can retarget those who engage your ad. If the images used align with your audience this can help alot.
3. Definitely consider layering your targeting interests, etc. Research similar apps and audience types. Combine them to be.. if user has liked X interest AND has liked Y interest (sometimes a 3rd ->) AND has liked Z interest. This will shrink your audience but narrow it down more to the core users.
4. Test LOTS of images/videos first, it's easier to work with a 6+ relevance score right off the bat than a 2.
5. Consider doing a different method.. write a blog post about how XYZ app is disrupting the space or someting or a blog post that asks a question.. "Is XYZ app about to overtake (insert recognizeable app name) in 2018". And build an audience profile off of that + create a lookalike.


Cheers,


Andrew

---

Another thing I'd do is to run an engagement campaign at the same time with a portion of the budget to the same post to:
1. add social proof - your ads look pretty lonely with 0 likes/comments even if you get clicks
2. get better traction with FB's algorithm and increase your RS

How are you redirecting the visitors to the app form FB? Through a pre-lander or straight to the app store?

---

Have you actually tested this? This is something I plan on testing within the next few weeks and I'm feeling very optimistic about it so I'm glad to see someone else mention it.

Was going to test both of these funnels:

1. Blog post / article > lander > offer
2. Blog post / article > offer

@OP How's it going anymore progress or updates?

---

Yes, I don't share information like this without having actully used it. This is not a new method, its well-known and works better for some offers/products/services than others. As for trying with an app, I have not.

But for Ecom/Shopify dropshipping we have gotten our 2nd lowest CPA's using this method on our hottest products, for traditional whitehat affiliate offers and for traditional businesses doing lead gen in regards to products/services this definitely works as well.

Ultimately, its a small initial investment you need to make but it helps to catch the attention of the right audience you need to be targeting, create lookalikes, etc. Once you have audience that you can make convert, then you have something to work with.



Andrew

---

This is great advice thanks for sharing with us. I didn't want to derail OP's thread so I sent you a PM.

Cheers.

---

Home > Paid Traffic Sources > Facebook & Instagram