Getting my 3rd, 4th and 5th tipped Teespring campaign (26)
What's up all!
A couple of weeks ago I posted about my first successful campaign on Teespring after my first week of trying it out here: http://stmforum.com/forum/showthread...profits-90-ROI
The week after I had a massively successful campaign as well that brought in $4.3k for less than $1.5k ad spend:
Now I'm faaaaar from being an expert because I do believe a huge element of my success so far was luck but I did learn quite a lot -- especially since this campaign was followed by a whole week with pretty much nothing working. I had a couple of campaigns that showed promise in the beginning, but eventually flattened out with zero sales and ended up being a waste of ad spend.
I'm posting this follow-along because I've finally broken through my cold streak last night and I decided to share the process here as typing things out helps me get a lot of mind clutter out of the way. I've been very guarded and quiet with my Teespring success so far, but I'm hoping that discussing it will open up new opportunities and good karma 
This is a mammoth of a post since I'm writing about 3 separate campaigns so far so I'll be separating them into different posts.
Oh yeah, and if anybody has a slot open with a Facebook and/or Teespring mastermind group, if you think I can be of some benefit I would love to be a part of it!
CAMPAIGN A
DAY 1
I have a few metrics that I keep close tabs on after launching a campaign. This way, I can predict whether or not it's going to be a success by the time that I've spent $5 to $10 on it. I’m not a believer of the commonly-preached tactic of spending $20 testing a shirt design -- that’s an easy way to get broke fast IMO!
CTR % is a decent metric but it doesn't tell the whole story. Take for example this campaign I was running earlier today:
One would think that this is good, but upon closer inspection:
(not sure why the second image says 15 actions while the one above says 14 post engagements but whatever)
I was only running the first ad at first. The numbers are deceiving; if you look closer, I was paying $1.32 per website click and had a 0.6% website CTR. If the profit per shirt sold is $10 (the normal profit for a one-color, $20 shirt with a minimum of 10 as its tipping point) I'd need a ~13% conversion rate just to break even.
Now this doesn't necessarily mean that it's a failed campaign (the engagement rate IS pretty high so at least I know that the message resonates with my target audience) but something must be tweaked. As you can see the second ad did much better in terms of cost per website click ($1.32 vs. $0.46) as well as website CTR (0.6% vs 1.6%).
Another tweak could also be the time of the day, so I'm planning to relaunch this again tomorrow at a different time and probably spend another $5 to $8 to see if it will make any sales then. If I can get a steady $0.50 website CPC here, I can break even with 1 sale every 20 clicks or a 5% conversion rate (which is pretty much the lowest conversion rate for all of my successful campaigns so far).
But just because a campaign has high website CPC doesn't mean it's a failure, because it COULD have a good conversion rate as you'll see in the next campaign.
CAMPAIGN B
DAY 1:
Now this campaign I'm super happy about because I did NOT expect it to gain any traction for reasons of my own. This campaign was on the verge of tipping while I was writing this post and basically tipped after 2 days of running ads ($130 earned so far).
I started this yesterday morning with a $10 budget targeting both males and females from 25 years old to any. When I start out campaigns, I usually bundle everyone together in one ad set. Not sure if this is the best way to go, but I want to see if I can get at least 1 conversion from the campaign from everyone -- I can then narrow it down later once it's made a sale or two.
I quickly got two sales a few hours after I launched this campaign, so I immediately increased the budget to $20. The day ended with 4 sales, and here's what the stats looked like at the end of the day:
Spent today: $15.09
Cost per website click: $0.94
Website CTR: 1.05%
Conversion rate: 25%
Cost per conversion: $3.77
I had four ads running, but because of the way Facebook displays ads it's pretty difficult to split-test ads since it uses optimized display (man, how I wish FB gives you the option to display all ads evenly within an ad set the same way that Adwords does!!). I paused the first ad and ran the second ad during the last few hours of the day, and it gave me these:
All 3 post engagements were website clicks, so not only was the cost per website click much lower at $0.40 but the website CTR % was also twice as much at 2.01%. Now, this is far from being statistically significant since the data is so small but I decided to take a leap of faith and run that ad exclusively for the second day.
Total Revenue: $40
Total Costs: ($15)
Total Profit: $25
DAY 2:
I was wary of increasing the budget because of the high costs. Only reason I kept it going was because it was converting really well, but I didn't want to go too loco with increasing the budget since I could easily lose money if for some reason the market doesn't buy on the next day as I've seen in the past.
I dug deeper into yesterday's data and decided to split my target audience into two demographics. I increased the daily budget into $7 for one group, and $15 for the second group -- pretty much a safe and proportionate increase in budget so that if it runs without making any sales the next day I wouldn't lose so much money.
Today, sales kept happening throughout the day, so I kept increasing/decreasing the budget accordingly. No real science to it -- I'd watch my stats every hour or so and as soon as one ad set brought in 2 sales, I'd increase it by $5 to $10. At the end of today I spent $30 and brought in 9 sales for $90 revenue.
Here's the ad set 1 performance:
Cost per website click: $1.41
Website CTR: 0.56%
Conversion rate: 20%
Cost per conversion: $7.04
(Horrible performance, as you can see!! But still somehow managed to make a small profit but definitely some tweaking required!)
And here's ad set 2 performance:
Spent today: $19.03
Cost per website click: $1.19
Website CTR: 1.71%
Conversion rate: 43.75%
Cost per conversion: $2.72
Total Revenue: $130
Total Costs: ($46)
Total Profit: $84
CAMPAIGN C
DAY 1
This campaign proved to be another surprise. My potential audience for this one isn’t huge either -- approximately 20k, which from my experience is a decent enough size for a campaign to tip if it’s tightly targeted.
Nothing too crazy here -- I had two ad sets going on that was separated by gender and both were targeting the same age range. I started out with a $5 budget for each with the intention of checking the stats by $7 or $8 spent to see whether I should continue. Only had 1 ad variation running for each segment.
Here’s the stats for the males at by end of the day:
Spent today: $5.49
Cost per website click: $0.61
Website CTR: 1.67%
Conversion rate: 11.1% (screenshot doesn't show it but there was 1 sale that came from this ad set)
Cost per conversion: $5.49
Here's the stats for the females:
Spent today: $10.86
Cost per website click: $0.33
Website CTR: 2.44%
Conversion rate: 9.09% (screenshot doesn't show it but there were 3 conversions from this ad set)
Cost per conversion: $3.62
I dug into the reports deeper. It doesn't seem like there are any major trends poking out of the male ad set, so I left that alone and increased the budget slightly to $7 for today. As for the women, I created 2 ad sets for different age groups -- one with a $4.00 budget, and the more profitable one with a $15 budget.
I've also left the ads running for longer than I should have because I want to figure out which times this audience is most active. As a result of this I had my fifth conversion logged technically on the next day, so that won't count as part of day 1's statistics.
Total Revenue: $40
Total Costs: ($16)
Total Profit: $24
This campaign is looking good so far. There's a ton of sharing (at least with the women) and in the shares I see there's a lot of them tagging their friends. So far that hasn't translated to much sales just yet, but I do have a feeling that the time of the day plays a huge part as to when this audience will convert (for some reason, this audience just seems to prefer to buy in the afternoon). As of right now, this campaign has 7 sales already -- 3 away from the tipping point.
Anyway that's about it for the 3 campaigns I have running for now! I just launched a new one a couple of hours ago but the ad hasn't started displaying yet but I'll also write about that tomorrow.
Great stuff man, thanks for contributing!
Teespring is definitely a difficult but lucrative nut to crack!
Great work there.
I have spent clode to $1k, and got around 700$ back. Still looking for that one campaign that will take off and cover up all these costs.
I have few questions :
1. What type of ads you use : Dark Posts - or Simple photo posts with link in description
2. Are you using partner Categories - as it's not available for international affiliates, we just have online spenders behavior.
Originally Posted by hangman
1. Log in to power editor as you normally would.
2. Make your ad as you normally would but don't upload.
3. Make sure you have absolutely no FB pages or chat systems open - nothing should be interacting with FB other than the power editor.
4. Connect to a US VPN.
5. Refresh the power editor page.
6. The partner categories will appear.
7. Do your dirty work, then disconnect from the US VPN.
8. Click upload changes.
Basically, partner categories are open to all advertisers but only contain US customer data.
So, if you are targeting the US you can use these. However, Facebook has, I assume due to an oversight, stopped these categories being shown in the power editor to non-US IPs. It is a very trivial IP-based block...
I have talked to a FB employee about this. They were adamant I should see the categories and were surprised when I told them about the US VPN workaround in the power editor.
They said it was fine for me to use that workaround for now - until FB realises they haven't enabled this properly.
Disclaimer: Be very very careful about using a VPN from a different country when toying with Facebook - you don't want to trigger suspicious activity. Hence why you must make absolutely sure you don't accidentally try to open FB pages, god forbid the ad manager, while using this workaround.
Originally Posted by zeno
hey i did this. Logged in by US IP - i did see partner categories, but what i meant was mainly purchase behavior - which i didn't see. So is there any workaround for that.
Great share Carlo!
Any particular reason you use 'post engagements' as opposed to 'conversions' as your goal on the ads??
Also, have you tried oCPM?
I've done well with Teespring (mid $xx,xxx from 0 in the past 4 months or so).
I've found that using conversions as the goal gives me a better view of things, also use oCPM exclusively now as it's worked out best, in my hands anyway. Also all ads are dark posts (for hangman).
Originally Posted by hangman
Not sure this is what you want! Can you be more specific with the behaviour you expected - have you seen it before? On another site e.g. blog site?
Originally Posted by zeno
Not sure this is what you want! Can you be more specific with the behaviour you expected - have you seen it before? On another site e.g. blog site?
Oh yes you're right - there's a hell of a lot of them missing based on other lists/images:
https://www.facebook-studio.com/news...geting-feature
http://www.ppchero.com/facebooks-par...e-list-so-far/
It's possible that FB has removed some partners or are omitting ones if your account isn't US-based.
I'll investigate further and find out what's going on.
There is definitely an element of luck in affiliate marketing. However, it depends on how you take advantage of that luck. You certainly seem like your on the right path to making some decent numbers. Keep scaling, and squeeze everything you can out of your angle/s!
What is stopping you from opening up your budgets and scaling at the moment?
Originally Posted by zeno
This is very important targetting factor which we non US affiliates are missing.
I've been using Website Conversions too and auto bidding which has worked well but it has been hard to keep CPC costs down if CTR drops in the slightest.
Originally Posted by hangman
I'm going to talk to a rep at a PMD about this, they should have an idea about what determines what is shown for an account.
Thanks for the update Carlo. I haven't had much luck with TeeSpring. I ran about 40 campaigns so far, but could only get 1 to tip. I'll keep trying until I figure it out. It's definitely a difficult but lucrative nut to crack as Zeno said 
Keep the updates coming!
I've been getting very frustrated with a lot of things about Facebook recently. The difficulty of setting up different ads for split-testing, the poor UX on their reporting, lack of dayparting (I've been pausing and unpausing my campaigns which I recently found out is a big no-no from one of the threads here), etc. So I signed up for a free trial over at AdEspresso and launched a couple of new campaigns from there. Hopefully everything works out -- will see real soon!
Will post an update on my current campaigns later or tomorrow -- been very pressed for time since I posted this! But good news is that while Campaign A failed, Campaign B and C have both tipped and are doing pretty well.
Originally Posted by hangman
2.) No partner categories or purchase behavior targeting -- I can only imagine it would be much more profitable if I did, but since I don't have a US-based advertising account I gotta make do with what I have so far. I tried zeno's tip, but I got the same results as you.
Originally Posted by flicker
Originally Posted by Mr Green
I do increase the budget in small increments as long as they make sense -- for example, if the budget is at $10/day and I make 2 sales, I increase that to $15 or even $20 depending how gutsy I'm feeling at that moment
Not really sure if it's the right strategy or if I'm leaving money on the table but so far it's working okay. Any thoughts would be appreciated! Originally Posted by zeno
Originally Posted by houssem
Keep the updates coming!
Whew!
Total revenue for these 3 campaigns: $1,109.79
Total spend for these 3 campaigns: $289.20, $91.83, and $185.33 = $566.36
For a total profit of: $543.43
Not bad, definitely not great by any means but I'm happy with it
Didn't get to update this thread as frequently as I wanted to in the past few days as I was busy working on the campaigns as well as doing personal shit. Which I know defeats the purpose of a follow-along, but what can you do...
I'm happy to report that my first month membership here at STM has been much more than worth it! I've learnt more about running PPC campaigns than ANY course has taught me, ever. Although I wasn't posting much, I had a couple of huge insights about Facebook PPC mostly thanks to zeno:
* I finally understood how Facebook campaign goals work -- I used to just put up Page Post Engagements and hope for the best there, but since I switched over to using Clicks To Website I've found increases in conversions. Not all the time, because there were instances when the PPE campaigns have a higher ROI than the CTW one. My next goal will be to split-test PPE vs. CTW with campaigns that have proven themselves.
* I learned about the difference in bidding strategies (basically using CPC bidding if your market tends to interact a lot because you'll benefit from "free" website clicks) which I'm looking forward to implementing in future campaigns.
* I'm going through the Newbie Guides over and over again and I'm learning something new everytime. It might not be directly about Teespring but it's up to me to apply the knowledge to my own campaigns. Learning about all of this is a wet dream for analytics junkies like me.
I do have a question regarding segmentation though: how big can an ad set segment get before I need to start thinking about more granular groups segmented by age?
I ask because I usually segment by interest and gender first. Before, I used to break down ad segments by a ton (25-34, 35-44, 45-54, 55 to Any) regardless of how many people would end up being targeted. What I do now is making my ad sets much larger (at least 30 to 50k reach) before I even think about breaking it down by age groups. So for example women who are 25 and older who like scotch tape, and then another ad set for men.
My question is, when is an ad group too large that you actually need to break it down by age groups? I know the answer to this question is "it depends" but some I'm interested in knowing some general guidelines that you might follow. For instance, I understand if you're promoting something that's more catered to young people vs. old people then yes you have to segment by age.
==========
I'm currently in the middle of another huge campaign. Don't have time to post about it since I'm busy scaling the thing but I'll report more in my next post!
Besides the obvious splitting when a campaign is likely to appeal to young and old differently, I would advise almost always splitting off 13-17 from the rest of the pack, and splitting the 18-65 group more tightly if your collective audience is above say ~100k.
In the end it's wise to just test this by running a broad campaign and running some smaller scale tests on the side with a high performance creative.
Originally Posted by thisiscarlo
Total spend for these 3 campaigns: $289.20, $91.83, and $185.33 = $566.36
For a total profit of: $543.43
Not bad, definitely not great by any means but I'm happy with it
Originally Posted by hangman
Originally Posted by zeno
Originally Posted by thisiscarlo
Others won't get white listed no matter what.
Originally Posted by hangman
Thanks for the suggestion zeno!! I'm hoping this will make an even bigger difference to my bottom line.
Originally Posted by hangman
To the best of my knowledge, these started out being available to US accounts only but were later rolled out - since there is no reason why a UK-based account shouldn't be able to use 3rd-party data on US users.
However, knowing FB, it's possible they leave these out unless your account is in good standing (i.e. say nope if you ask).
Home > Facebook & Instagram > Follow-along Campaigns