Hello everyone, 👀
I understand that the last quarter of the year is a busy and important time for every business. In 2021, you’re most likely trying to execute plans while navigating a sea of platform and privacy changes across the entire online advertising industry (but mostly on Facebook Ads lol)
I know that, you know that…
Overall, we know that ESPECIALLY NOW, most businesses are trying to take over new markets or expand to their current ones, and today’s case study is about taking over a new country, with a new pixel and a small budget, because this is usually how it goes in the beginning. This case study will answer a few questions:
  • How to start a new ad account with 0 data in a foreign country
  • How to approach things when you don’t know what sells
  • How to approach things on a smaller budget
Before we move on, I would like to mention a couple of IMPORTANT details about the accounts for you to have some context:
  • This account is in Australia, a beauty niche (in a pretty competitive sub-niche as well).
  • ROAS target: 3, Target CPA: $35, but our focus was ROAS as per our discussion with the client.
  • We started with a smaller budget to test out the waters, $100 / day.
  • The pixel was “learnt” from 0, and from the first days, we were able to identify what creatives and products worked, and what type of audiences matched those creatives. This way, it was just a matter of trial and error until we were able to stabilize this campaign.
Let’s start with the “beginning”. In the below screenshot you have attached data from both the month of August, as well as the month of September when we stabilized this account. Most data is coming from COLD targeting, and after investing some resources (both time and money), we were able to start retargeting as we had customers in the pipeline.
First 5 days of this account, we launched campaigns with a mix of conversions, from view content, to add to cart. Because we already knew the metrics expected (estimated costs per different actions in the funnel), we didn’t need more than 5 days to see that things are working in our favour and we moved our optimization down the funnel, to initiate checkout and then to purchase. In these 5 days, we were able to get 10 purchases, and ROAS was pretty good, standing at a whipping 7.09, and a CPA of $25.79.
For these first 5 days we used the following strategy with the lowest cost:
1 x ABO Campaign optimized for view content with 5 ad sets and 5 ads. ($5 / ad set)
1 x ABO Campaign optimized for add to cart with the same audiences, and 5 ads. ($5 / ad set)
1 x ABO Campaign optimized for initiate checkout with the same set up mentioned above.  ($10 / ad set)
Creatives used in the first 5 days: Dynamic Catalogues & Manual Catalogs. Remember, we did not know exactly what sells the most so we had to test the waters before we came up with other types of ads. This business had over 150 products in stock.
Pro tip: We used different variations of catalogs: product sets, all product catalogs, catalogs with custom frames and catalogs with NO buttons, NO primary text (You might think I’m crazy ain’t it?) – BUT THOSE WORKED BEST, let’s be honest: They don’t really look like ads.
Great, after these first 5 days we have identified that product sets with no buttons & no text convert best and we identified 3 winning audiences. We looked inside the store analytics to see what people were looking for and we created new product set combinations to test out, by introducing them into those 3 campaigns.
Other findings:
  • Manual catalogs with product sets worked better than “all products” catalogs, which made sense for us as the pixel did not have data to know what products sell.
  • Catalogs with no text and no buttons delivered the best CTR as well as the best cost per result.
  • We split tested collection vs product landing page and we noticed that the (surprisingly weird) collection page worked better than specific product URL.
A very important aspect to mention here is that we didn’t look just at ROAS and CPA. Instead, we focused on CTR & CPC to make sure we won’t have surprises in the near future. We inspected the ads constantly and compared the last 3 days with the other 3 days before that.
The first campaigns (optimized on view content and add to cart) were stopped as we started focusing on initiating checkout & purchase.
After these 5 days we were confident in our findings and switched our focus from ABO to CBO, by stacking the best audiences with the best creatives. (PS: We created a seed campaign to store the winning ads with post id where that applied).
For the following 10 days, we introduced new but similar product catalogs and we started testing new audiences as well in an ABO campaign, so to recap, we created:
1 x CBO with the winning audiences, 3 ad sets, and best creatives ($65 / campaign)
1 x ABO with new audiences, 5 ad sets, and 5 new catalog variations with collection links ($5 / ad set)
This allowed us to test out new assets, and stabilize the existing ones, from here on. As soon as we gained more traction and the other 10 days passed, we started creating manual ads as well, for individual products.
One more thing we added here is rules (not too conservative though), to keep costs under control, especially since we used mainly lowest cost.
To recap:
We started small, with just $100 / day, “teaching” the pixel what are we expecting from it in terms of costs, optimizing for cheaper actions like view content, add to cart and initiate checkout.
We started with ABO and then continued with CBO. Every time we created a new test, we either created another ABO or a CBO with minimum spending to enforce equity. After a month and a half, we spent 4.9k and got back 19k, which led to a ROAS of 3.87 with a CPA or $39.88.
I know that is a rough period to start advertising, especially with 0 data but you can always ask yourself the following questions:
  • Did I test out all types of creatives available?
  • Did I optimize for different types of events?
  • Are my up-the-funnel costs within targets?
  • Did I pay attention to other metrics, rather than ROAS and CPA?
  • Did I try different angles for my ads?
  • Is my funnel doing the heavy lifting? Like the upsells, downsells, follow-ups, and re-engaging with the client?
In case you tried all the above and things are still not working, don’t beat yourself up, use other ways of promoting your business. ALWAYS use an omnichannel strategy to increase your chances of success.
omnichannel campaign
resource centre1
resource centre 2

Leave a Reply