Home > Facebook & Instagram > Follow-along Campaigns

Facebook + Sweep [ Stickupkid Style :) ] (16)


06-10-2018 04:42 PM #1 itzpeter (Member)
Facebook + Sweep [ Stickupkid Style :) ]

Hi everybody, I was inspired by stickupkid's pin submit game, so I'm deciding to give it a try. Because it's the weekend, I couldn't get any pin submit offers approved yet, so i'm testing this on a department store gift card offer SOI style. I started campaign about 12 hours ago and this is what I have so far.

GEO: Tier 1
Offer: Win gift card
Offer Payout: 3.10 CPA
Offer style: Single Opt In


Tools:
1) Ad Account | I warmed it up with about $150 spend. It got billed about 2 times before I started this.
2) Clean non-aggressive 4 question survey lander on fresh domain.
3) Ad-network

Tools that are not used:
1) cloaker
2) tracker

Alright lets go:

step 1) I searched images and gathered 6. Images like: Outside store, Inside store, Long lines, People paying at cashier, shopping bag, shopping cart, etc. The point is to capture different directions, angles, and styles, since different people respond differently to images.

step 2) Created a relevant FB page. I went pretty aggressive here. example, if my offer was win Macy's gift card -> Then my page would be Macy's Department Store or something like that.

step 3) I created 6 adsets - each with same targeting / Budget for each ad set was $15.

step 4) I created 6 ads - each with same headline copy / image is different for each ad. [testing images first]

Facebook Stats:



Adnetwork stats:

Clicks: 73
Conversions:11
Payout: $34.10

Conclusion:

Cost: $37.39
Revenue: $34.10
Profit: -$3.29
ROI: -8.8%

Learnings:

Stickupkid is right: you don't need a cloaker or tracker for facebook sweeps. When I submitted that 4 question survey lander and it was under review, my heart was going crazy until it was approved [ and I started popping champagne ]

Also, people give positive comments - it feels good doing non-aggressive

Moving Forward:

I think I'm going to test more images to find a winner. Then, start testing more angles. Currently it looks like Ad 5 & Ad 4 has the highest CTR so far. But, I'm going to let it run a little longer for stats to make a firm conclusion.

Questions:

1) I noticed my relevance score on the ads is 6. How do I improve that?

2) Also, I have no idea how to tell which ad is making the conversions. And does it even matter to see which ad is making conversions or should I just choose the one with the best CTR and lowest cost and just be blind on the conversions. ??


This is my first follow along on the forum I think. I would like to thank everyone who took their time to read through this and I would also appreciate any tips or feedback. Oh and special thanks to Stickupkid, for this would not be possible without him.

​*Cheers Mate*


06-10-2018 05:13 PM #2 stickupkid (Senior Moderator)

First of all great results, if you ditch ad 1,2,6 you probably already in green man, that's great! Another great example CTR is so important to lower your CPC.

130 clicks from FB and 73 offer clicks seems like a good CTR on the prelander.

1. What's your exact targeting, and which placements did you choose? Try to target [brand] only (if audience is big enough), and avoid competitors or similar shops at the start.

2. I would normally say yes (to track where conversion come from), but I guess your text, targeting etc is the same, only different images. IMHO I can't imagine an image of a grocery bag would bring in more conversions than a store front of that certain brand. I always optimize on ad level, to get the lowest cpc and let those run.

Try not to be too active on ad changes, duplicates and budgets at the start, you are still on thin ice with sweeps so try to avoid attention. Facebook doesn't like sweeps too much, although your ads have good feedback as you mentioned.


06-10-2018 05:14 PM #3 trafficbae (Member)

Hey @itzpeter, thank you for share!
How did you warmed up account? For what you'd spend $150 - it was something white hat?


06-10-2018 05:38 PM #4 itzpeter (Member)

@stickupkid

Hey thanks for the feedback. I'll take them out immediately and see what happens.

1) my targeting is women 25+ | interest of shopping and fashion | broad with 4-5M peeps. I couldn't do brand only because audience size is too small. Placements I do mobile only & Newsfeed only.

Hmm I see, thanks for the advice. and yeah your right haha the grocery bags seem to not do so well. So, to avoid detection, I should let this run for a little bit and get billed couple times before I start making changes to the headline text right?

If I wanted to track which adset is giving me conversions. how do I do it?


06-10-2018 05:41 PM #5 itzpeter (Member)

@trafficbae

Ok so this is how I do it haha. Don't get mad at me if it doesn't work for you. I just use a generic image and do an Ecommerce type Ad linking to an Amazon product page.

but moving forward, clickdealer just released dropshipping whitehat offers. So, I would probably run those to warm up accounts and make back some money at the same time

So i just do something like 25 bucks a day for 6 days.


06-10-2018 08:06 PM #6 diamond (Member)

Great results so far!

Usually during the first two days is when you get the highest costs and then it starts to settle down. Often you will also see the relevance score go up by then.

I wouldn't get to any definite conclusions yet as you've only run this for 12 hours. FB's algorithm needs about 3 days to fully optimize to get you the best bang for buck.


06-11-2018 12:35 AM #7 servandosilva (Member)

It’s important for sure to know which ads are bringing conversions.

Clicks and CTR are nothing if they don’t convert.

For all you know at the moment your conversions could be coming from the lower CTR ads or one from each or all from just one ad and you’re about to pause them.

Is the 4 question survey hosted in just one URL or distributed in 4-5 URLs (one for each question)?


Sent from my iPhone


06-11-2018 03:49 AM #8 itzpeter (Member)

@servandosilva

I see, totally agree with you. Can you please show me how I can track it?

The 4 question survey is hosted in just one URL


Sent from my iPhone using STM Forums


06-11-2018 08:10 AM #9 stickupkid (Senior Moderator)

Quote Originally Posted by itzpeter View Post
@stickupkid

Hey thanks for the feedback. I'll take them out immediately and see what happens.

1) my targeting is women 25+ | interest of shopping and fashion | broad with 4-5M peeps. I couldn't do brand only because audience size is too small. Placements I do mobile only & Newsfeed only.

Hmm I see, thanks for the advice. and yeah your right haha the grocery bags seem to not do so well. So, to avoid detection, I should let this run for a little bit and get billed couple times before I start making changes to the headline text right?

If I wanted to track which adset is giving me conversions. how do I do it?
1) That's why relevance score is 5/6, targeting is a bit too broad now. How big will the audience be if you target brand only? If it's 100k + it's fine to test images first to see which do best. After that you can set a broader audience on the best performing images. It's likely they will have a good CTR on broader audience too.

"Hmm I see, thanks for the advice. and yeah your right haha the grocery bags seem to not do so well. So, to avoid detection, I should let this run for a little bit and get billed couple times before I start making changes to the headline text right?"

What I mean that the IMAGE is mainly for ATTRACTION, and doesn't have any (or very very little) impact on conversion rates. The prelander has purpose to create desire and action. So if you do a shopping bag or a store-front, doesn't effect CR in general at all. That's why it's not really neccesary to place a pixel and see where conversions come from. I must say it's a strategy I prefer, but doesn't say it's the best.

If your ads are in different campaigns / ad sets with different settings, yes then it's needed to place a pixel and see where conversions come from.


06-11-2018 09:53 AM #10 diamond (Member)

It depends on how you want to track the conversions.

If you're running a traffic campaign, I'm not too sure, but the conversions tracked with a pixel will just be for reporting purposes. However, the pixel will still be gathering data which you could be using at a later time for custom audiences / LAAs. I don't really recommend using a pixel if you can't actually track a lead firing on the advertiser end. Conversion campaigns typically result in a way higher cost per result and it will eat your ROI if you don't even feed it the juicest data it can get.

If you don't care for the pixel, you can give each ad it's own URL parameter value like "ad=ad1", "ad=ad2", ad=ad3", etc. All clicks going to the destination link for ad 1 will result in "https://destination-url.com/?ad=ad1". You can read this parameter with JavaScript and insert it into a sub ID of your offer URL. The offer URL would then become "https://offer-url.com/?u=12345&o=54321&sub5=ad1". Now whenever you generate a new lead, this value will be tracked by your network and you'll find it in your conversions report to see which ad performs best as far as lead conversions go.

You could also run both solutions simultaneously for maximum tracking and if you want to spot any discrepancies, etc.


06-11-2018 10:00 AM #11 stickupkid (Senior Moderator)

Thanks for the addition @diamond! I must admit I am not using any tools since I am not a techy, and way too lazy. Besides that I don't want to make FB smarter about the campaigns I run. Again, it might not be the best way, but it works for me eventually.


06-12-2018 03:36 AM #12 itzpeter (Member)

@diamond

dude bro thank you so much for that! I am not techy at all so it's gunna take me some time to dial all that in.


06-12-2018 03:43 AM #13 itzpeter (Member)

Quote Originally Posted by stickupkid View Post
1) That's why relevance score is 5/6, targeting is a bit too broad now. How big will the audience be if you target brand only? If it's 100k + it's fine to test images first to see which do best. After that you can set a broader audience on the best performing images. It's likely they will have a good CTR on broader audience too.

"Hmm I see, thanks for the advice. and yeah your right haha the grocery bags seem to not do so well. So, to avoid detection, I should let this run for a little bit and get billed couple times before I start making changes to the headline text right?"

What I mean that the IMAGE is mainly for ATTRACTION, and doesn't have any (or very very little) impact on conversion rates. The prelander has purpose to create desire and action. So if you do a shopping bag or a store-front, doesn't effect CR in general at all. That's why it's not really neccesary to place a pixel and see where conversions come from. I must say it's a strategy I prefer, but doesn't say it's the best.

If your ads are in different campaigns / ad sets with different settings, yes then it's needed to place a pixel and see where conversions come from.
yeah its a small audience size, I think it was like 200k-300k. not a really popular brand.


06-12-2018 06:55 AM #14 stickupkid (Senior Moderator)

Quote Originally Posted by itzpeter View Post
yeah its a small audience size, I think it was like 200k-300k. not a really popular brand.
That is actuallly a decent audience to test your first ads honestly. This way you filter best ads and when the 200-300k audience is milked out you can duplicate the best campaigns on a broader audience!

In meantime find more more more images and launch some new ones daily/weekly to find the sweetest spot!


02-21-2019 01:28 PM #15 gritaction (Member)

hello peter, i saw ur video interview with amy in fb! very interested to know how does ur campaign go so far?
did u manage to use any tracker to track eventually?


02-21-2019 03:25 PM #16 stickupkid (Senior Moderator)

Peter by the way is a real good guy, awesome to talk about his progress once in a while.


Home > Facebook & Instagram > Follow-along Campaigns