Scaling Ando to 3.5 Meta Ads ROAS in 30 Days

Scaling Ando to 3.5 Meta Ads ROAS in 30 Days

Scaling Ando to 3.5 Meta Ads ROAS in 30 Days

Scaling Ando to 3.5 Meta Ads ROAS in 30 Days

Service

Service

Paid Media

Paid Media

Quick Stat

Quick Stat

3.5x ROAS

3.5x ROAS

Year

Year

14 May, 2024

14 May, 2024

Introduction

Ando is the world’s first gummy designed to prevent Alcohol Flush, a symptom often called “Asian Glow” that causes redness and an elevated heart rate when drinking Alcohol. Ando is also the first brand completely owned and launched by Flighted. We used our playbook from day one, controlling everything from content sourcing to landing page optimization and messaging testing.

We launched Ando on April 7th, hoping to get the brand to profitability via Meta ads with just $4,000 in ad spend over the course of a 30-day sprint. A goal this ambitious required 3 things:

  • A high volume of diverse creative (in our case, roughly 20 concepts and 50+ total ads)

  • A detailed knowledge of our customer’s pain points and winning messaging

  • High-impact price, offer, and landing page testing

Content Sourcing

We knew that having a diversity of creative formats would be key to our success. By spreading our chips across a variety of content types - UGC videos, more polished explainer ads, founder story videos, organic tiktoks, static ads, before-and-afters, etc. - we would increase our likelihood of finding creative winners with such a limited budget.


We used a combination of the TikTok Creator Marketplace and UGC sourcing platforms like Brands Meet Creators to find cost-effective content creators who understood our product and were willing to get creative with storytelling and messaging. The best video ads don’t feel like ads!

We compiled a list of the top performing static ad formats using Foreplay, knowing that these would be a low-cost way to get quick insights into what messaging would work best. We leveraged a wide variety of formats. Two of our favorites were “Old Me / New Me” (a take on the classic before-and-after format) and the Tweet Style UI ad.

We created a high volume of traditional explainer videos that tested a high volume of different hooks. We used Eleven Labs to quickly generate natural-sounding voice overs using AI that allowed for easy iteration of messaging, while still making the creative feel natural, as if it was made by a content creator.

Lastly, we worked with the brand’s co-founder to make more heartfelt“founder story” videos where he simply explained why he started the company after facing the issue of Alcohol Flush himself. We knew this ad format would be key to whitelist from his personal social handle later, driving up Click-Through Rate (CTR) and Thumbstop Rate (TSR). 


Launch Strategy

Putting this all together and testing it in the most cost-effective way was going to be the most critical step of our launch. We had to balance finding statistically significant learnings on what was working with minimizing the amount of budget we spent on any single test. In order to do this, we relied heavily on Meta’s machine learning to dynamically allocate spend across creative concepts it thought would work, with minimal amounts of ad spend. We structured our ad account as follows:

Campaign 1: Video ads - $80 per day - Campaign Budget Optimization - Broad targeting - Highest Volume bidding - One  Dynamic Creative ad set for each creative concept (no more than 8 ads per adset)

Campaign 2: UGC ads - $80 per day - Campaign Budget Optimization - Broad targeting - Highest Volume bidding - One Dynamic Creative ad set for each UGC creator

Campaign 3: Static ads - $80 per day - Campaign Budget Optimization - Broad targeting - Highest Volume bidding - One Dynamic Creative ad set for each static concept

We knew that we needed to separate static ads in the early stage of testing, as they often don’t perform well when grouped together with video ads. Similarly, we put UGC in a separate campaign from our traditional video ads so we could better control budget across these different ad concepts.

We leveraged Campaign Optimization to quickly sort through which of our 50+ ad concepts Meta thought would actually perform, and chose not to use bid caps because we knew that  none of our campaigns would spend with zero spend history in a brand new ad account. We also knew broad targeting historically works best with new ad accounts - niching into interests too early can send Meta a false signal, and Meta optimizes extremely quickly on broad targeting. Ultimately, your creative does the targeting.

Once our ads went live, we began aggressively reverse-optimizing. This is what most advertisers get wrong in a new ad account. They start consolidating spend behind ad concepts that work far too quickly, counting out the concepts that have yet to get spend. Given that we were severely budget constrained (an $80 campaign budget meant we were spending less than $5 per ad per day), we quickly paused the ad sets that Meta first allocated spend to so that we could rapidly test through the rest of the concepts in the campaign. Doing otherwise would have meant burning through our low $4k budget far too quickly. We wanted to get to the consolidation phase as quickly as possible. 


Consolidating and Optimizing

Once we had a decent signal on each creative concept/ad set (roughly $50 spend, minus the adsets that were immediately deprioritized by Meta’s delivery algorithm), we launched phase 2: creative winner consolidation and landing page testing.

The consolidated ad account structure was extremely simple. We took the Post IDs of our top 10 ads from the previous round of creative testing to ensure that we didn’t lose social proof or Meta’s high Estimated Action Rate attributed to each of those ads, and launched them into 2 ad sets: a broad targeting adset and a lookalike of users who engaged with our previous round of ads + 180 day website visitors. We leveraged Campaign Budget Optimization to quickly find a winner between these two audiences. Our primary goal at this point was to exit the learning phase by generating 50 conversions on this winning ad set in a 7 day period. 

Simultaneously, we knew that we wouldn’t have enough traffic to run more than one statistically significant landing page test, so we made sure to test a completely unique landing page format to increase the test’s likelihood of being high impact. We tested our shoppable homepage against a completely different, more advertorial problem/solution landing page that twisted the knife a bit more on the pain points of turning red while drinking. We used VWO to run this A/B test as a URL redirect so that we didn’t need to edit our ad URLs on Meta and re-enter the learning phase. 


Iterating and Scaling

At this point, we were consistently hovering around a 1.5 ROAS, breakeven roughly 50% of the time. We needed to do 3 things to push these campaigns into true profitability and scale our ad spend: iterate on winning messaging, utilize whitelisting, and launch Meta’s Advantage+ Shopping campaign type.

We iterated on messaging by examining the commonalities of the winning ads from our first 2 rounds of testing, AND by leveraging post-purchase survey data we had started collecting several weeks prior. We saw the phrase “turn red when you drink” outperform the more clinical “Alcohol Flush”, for example. Post purchase survey data showed us that most customers were using Pepcid to mitigate Alcohol Flush. We took these learnings and applied these hooks to our top-converting ads, leveraging angles that attacked Pepcid directly. Our theory was that these high converting ads, when combined with more intriguing/higher CTR hooks, would perform much better.

Whitelisting was critical to our success. It’s very common to see extremely high CPCs when you launch a new ad account in the supplement industry in particular. Whitelisting is the #1 way to bring down CPCs by running ads from the creator’s handle instead of your brand handle. We reached out to the UGC creators whose ads saw the best performance in Round 1, and negotiated whitelisting access to re-launch these ads in a format that would drive cheaper clicks. 

Lastly, we waited until the very last second (once we had spend about $3k of our $4k budget) to launch Meta’s incredibly powerful AI targeting campaign format, the Advantage+ Shopping Campaign (ASC). The reason we waited so long to launch this is that ASCs are useless unless they are trained on historical conversion data, just like any other machine learning model. Launching this too early would have wasted ad spend and not been effective, so we waited until we had driven more than 100 conversions in-platform. Another advantage of ASCs is that they can handle a much higher volume of ad creative than a typical ad set, so we were able to combine both our historical winners with our new iterative tests into a single adset.  

Breaking Through to Profitability

At this point, everything was finally coming together. We saw some of our iterative winners begin picking up major traction, particularly the Pepcid alternative angle. We put a bid cap on our campaign with Broad targeting once it exited the learning phase, ensuring it only spend efficiently. Our Advantage+ campaign was seeing 40% greater ROAS than the broad targeting campaign. 

This is another point where many advertisers mistakenly continue to optimize, keeping their ad account in the learning phase. We saw ROAS increasing at this point and knew  it was time to take our foot off the gas, reduce our rate of changes in the ad account, and let Meta do the work. We paused our landing page test and pushed 100% of our traffic to our winning landing page, the advertorial style problem/solution page. Conversion rate grew.

After 3-5 days of inactivity, a few of our iterative winners began really taking off. ROAS climbed to 2x, and even 3-4x in our Advantage+ campaign. Our whitelisting ads also started seeing traction. We quickly started raising budgets by the max 15% per day that wouldn’t trigger the learning phase, and within a few days we were spending $600+ per day at a 3x ROAS! 

Conclusion

By the time we were able to scale Ando to meaningful profitability, we unfortunately were on track to sell through the company’s entire first Purchase Order, so we were forced to pull back on ad spend. You can see the full “hero’s journey” of our Meta Ads sprint in the below graph:

The quick success of Ando was a validating exercise that showed how powerful Meta can be with effective media buying strategy and data-driven creative testing. 

Introduction

Ando is the world’s first gummy designed to prevent Alcohol Flush, a symptom often called “Asian Glow” that causes redness and an elevated heart rate when drinking Alcohol. Ando is also the first brand completely owned and launched by Flighted. We used our playbook from day one, controlling everything from content sourcing to landing page optimization and messaging testing.

We launched Ando on April 7th, hoping to get the brand to profitability via Meta ads with just $4,000 in ad spend over the course of a 30-day sprint. A goal this ambitious required 3 things:

  • A high volume of diverse creative (in our case, roughly 20 concepts and 50+ total ads)

  • A detailed knowledge of our customer’s pain points and winning messaging

  • High-impact price, offer, and landing page testing

Content Sourcing

We knew that having a diversity of creative formats would be key to our success. By spreading our chips across a variety of content types - UGC videos, more polished explainer ads, founder story videos, organic tiktoks, static ads, before-and-afters, etc. - we would increase our likelihood of finding creative winners with such a limited budget.


We used a combination of the TikTok Creator Marketplace and UGC sourcing platforms like Brands Meet Creators to find cost-effective content creators who understood our product and were willing to get creative with storytelling and messaging. The best video ads don’t feel like ads!

We compiled a list of the top performing static ad formats using Foreplay, knowing that these would be a low-cost way to get quick insights into what messaging would work best. We leveraged a wide variety of formats. Two of our favorites were “Old Me / New Me” (a take on the classic before-and-after format) and the Tweet Style UI ad.

We created a high volume of traditional explainer videos that tested a high volume of different hooks. We used Eleven Labs to quickly generate natural-sounding voice overs using AI that allowed for easy iteration of messaging, while still making the creative feel natural, as if it was made by a content creator.

Lastly, we worked with the brand’s co-founder to make more heartfelt“founder story” videos where he simply explained why he started the company after facing the issue of Alcohol Flush himself. We knew this ad format would be key to whitelist from his personal social handle later, driving up Click-Through Rate (CTR) and Thumbstop Rate (TSR). 


Launch Strategy

Putting this all together and testing it in the most cost-effective way was going to be the most critical step of our launch. We had to balance finding statistically significant learnings on what was working with minimizing the amount of budget we spent on any single test. In order to do this, we relied heavily on Meta’s machine learning to dynamically allocate spend across creative concepts it thought would work, with minimal amounts of ad spend. We structured our ad account as follows:

Campaign 1: Video ads - $80 per day - Campaign Budget Optimization - Broad targeting - Highest Volume bidding - One  Dynamic Creative ad set for each creative concept (no more than 8 ads per adset)

Campaign 2: UGC ads - $80 per day - Campaign Budget Optimization - Broad targeting - Highest Volume bidding - One Dynamic Creative ad set for each UGC creator

Campaign 3: Static ads - $80 per day - Campaign Budget Optimization - Broad targeting - Highest Volume bidding - One Dynamic Creative ad set for each static concept

We knew that we needed to separate static ads in the early stage of testing, as they often don’t perform well when grouped together with video ads. Similarly, we put UGC in a separate campaign from our traditional video ads so we could better control budget across these different ad concepts.

We leveraged Campaign Optimization to quickly sort through which of our 50+ ad concepts Meta thought would actually perform, and chose not to use bid caps because we knew that  none of our campaigns would spend with zero spend history in a brand new ad account. We also knew broad targeting historically works best with new ad accounts - niching into interests too early can send Meta a false signal, and Meta optimizes extremely quickly on broad targeting. Ultimately, your creative does the targeting.

Once our ads went live, we began aggressively reverse-optimizing. This is what most advertisers get wrong in a new ad account. They start consolidating spend behind ad concepts that work far too quickly, counting out the concepts that have yet to get spend. Given that we were severely budget constrained (an $80 campaign budget meant we were spending less than $5 per ad per day), we quickly paused the ad sets that Meta first allocated spend to so that we could rapidly test through the rest of the concepts in the campaign. Doing otherwise would have meant burning through our low $4k budget far too quickly. We wanted to get to the consolidation phase as quickly as possible. 


Consolidating and Optimizing

Once we had a decent signal on each creative concept/ad set (roughly $50 spend, minus the adsets that were immediately deprioritized by Meta’s delivery algorithm), we launched phase 2: creative winner consolidation and landing page testing.

The consolidated ad account structure was extremely simple. We took the Post IDs of our top 10 ads from the previous round of creative testing to ensure that we didn’t lose social proof or Meta’s high Estimated Action Rate attributed to each of those ads, and launched them into 2 ad sets: a broad targeting adset and a lookalike of users who engaged with our previous round of ads + 180 day website visitors. We leveraged Campaign Budget Optimization to quickly find a winner between these two audiences. Our primary goal at this point was to exit the learning phase by generating 50 conversions on this winning ad set in a 7 day period. 

Simultaneously, we knew that we wouldn’t have enough traffic to run more than one statistically significant landing page test, so we made sure to test a completely unique landing page format to increase the test’s likelihood of being high impact. We tested our shoppable homepage against a completely different, more advertorial problem/solution landing page that twisted the knife a bit more on the pain points of turning red while drinking. We used VWO to run this A/B test as a URL redirect so that we didn’t need to edit our ad URLs on Meta and re-enter the learning phase. 


Iterating and Scaling

At this point, we were consistently hovering around a 1.5 ROAS, breakeven roughly 50% of the time. We needed to do 3 things to push these campaigns into true profitability and scale our ad spend: iterate on winning messaging, utilize whitelisting, and launch Meta’s Advantage+ Shopping campaign type.

We iterated on messaging by examining the commonalities of the winning ads from our first 2 rounds of testing, AND by leveraging post-purchase survey data we had started collecting several weeks prior. We saw the phrase “turn red when you drink” outperform the more clinical “Alcohol Flush”, for example. Post purchase survey data showed us that most customers were using Pepcid to mitigate Alcohol Flush. We took these learnings and applied these hooks to our top-converting ads, leveraging angles that attacked Pepcid directly. Our theory was that these high converting ads, when combined with more intriguing/higher CTR hooks, would perform much better.

Whitelisting was critical to our success. It’s very common to see extremely high CPCs when you launch a new ad account in the supplement industry in particular. Whitelisting is the #1 way to bring down CPCs by running ads from the creator’s handle instead of your brand handle. We reached out to the UGC creators whose ads saw the best performance in Round 1, and negotiated whitelisting access to re-launch these ads in a format that would drive cheaper clicks. 

Lastly, we waited until the very last second (once we had spend about $3k of our $4k budget) to launch Meta’s incredibly powerful AI targeting campaign format, the Advantage+ Shopping Campaign (ASC). The reason we waited so long to launch this is that ASCs are useless unless they are trained on historical conversion data, just like any other machine learning model. Launching this too early would have wasted ad spend and not been effective, so we waited until we had driven more than 100 conversions in-platform. Another advantage of ASCs is that they can handle a much higher volume of ad creative than a typical ad set, so we were able to combine both our historical winners with our new iterative tests into a single adset.  

Breaking Through to Profitability

At this point, everything was finally coming together. We saw some of our iterative winners begin picking up major traction, particularly the Pepcid alternative angle. We put a bid cap on our campaign with Broad targeting once it exited the learning phase, ensuring it only spend efficiently. Our Advantage+ campaign was seeing 40% greater ROAS than the broad targeting campaign. 

This is another point where many advertisers mistakenly continue to optimize, keeping their ad account in the learning phase. We saw ROAS increasing at this point and knew  it was time to take our foot off the gas, reduce our rate of changes in the ad account, and let Meta do the work. We paused our landing page test and pushed 100% of our traffic to our winning landing page, the advertorial style problem/solution page. Conversion rate grew.

After 3-5 days of inactivity, a few of our iterative winners began really taking off. ROAS climbed to 2x, and even 3-4x in our Advantage+ campaign. Our whitelisting ads also started seeing traction. We quickly started raising budgets by the max 15% per day that wouldn’t trigger the learning phase, and within a few days we were spending $600+ per day at a 3x ROAS! 

Conclusion

By the time we were able to scale Ando to meaningful profitability, we unfortunately were on track to sell through the company’s entire first Purchase Order, so we were forced to pull back on ad spend. You can see the full “hero’s journey” of our Meta Ads sprint in the below graph:

The quick success of Ando was a validating exercise that showed how powerful Meta can be with effective media buying strategy and data-driven creative testing. 

Ready to talk?

Book A Call

We are a growth marketing agency based in Brooklyn, NY.

Flighted

Ready to talk?

Book A Call

We are a growth marketing agency based in Brooklyn, NY.

Flighted