Skincare Brand


This case study is about a direct to consumer skincare brand based in the USA, who we’ve added an extra $1.6M in sales to their bottom line.

To respect our client’s confidentiality and trade secrets, we’re unfortunately not able to share the client’s brand name, creatives & ad copy.

They provide botanical, cruelty-free and vegan facial serums for 35+ women to improve skin appearance and to reduce wrinkles.

The owner of the brand was running the ads himself and wanted to improve his ROAS, scale his account and free himself to focus on more important tasks for his business.

His ROAS and revenue started to decline and managing the Facebook ads switched from being a growth driver to a bottleneck for his business.

The month before working with us, the account performance was at $35K per month ad spend at breakeven. Within 6 months we’ve been able to scale the account to $500K per month in revenue at almost 2X ROAS, which is now profitable for the brand and above our account-wide target of 1.6.

With an increase in new customers and repeat customers, we’ve been able to help grow his brand to almost $600K/mo in revenue.

So how did we turn his account around, triple the revenue and increase ROAS?

Step 1: Analyze

Before we take on a client at Sell Through Digital, we analyze the account and figure out – what are the exact bottlenecks that are causing a poor ROAS and stopping the account from scaling? Here were a few key issues discovered in this skincare brand’s account.

1) More creative testing was needed

Creative is the most important part of an eCommerce ad. Yet this wasn’t being consistently tested and focused on.

2) Overspend on ads with ROAS below target ROAS.

Since the client was manually turning ads on and off, there were leaks in the account. Multiple ads that continued to spend, despite the ROAS being below target.

3) Messaging could be improved throughout the customer journey.

Showing the same remarketing message to the same audience isn’t likely to address their needs. Different people have different objections as to why they buy, so a variety of messages need to be shown.

4) Website conversion rate could be improved.

Conversion rate optimization (CRO) is the process of improving the conversion rate and revenue from a website. The brand’s initial CRO had bottlenecks hurting conversion rate. Such as:

– Poor mobile user experience. For example, the menu wasn’t placed where it should have been and there was an intrusive pop-up which hurt user experience.

– Slow site speed and instability. The initial website was slow, drop off rate was high, the site crashed at times, and slowed down even more when increasing spend and traffic. We advised removing many redundant and bandwidth draining plugins within the site.

Our in-house developer and CRO expert Josef is now working on transitioning the site to a headless build, so that all the pages will load quickly and as we scale up the spend, it won’t slow down and hurt conversion.

5) Single product offers. AOV could have been improved.

Most of the offers were packaged as single products. There was no bundling of products to increase AOV and ROAS.

Step 2: Strategize

Here’s how our strategy and approach differed and thus scaled the account.

Firstly, we got clear on the ROAS targets that we need to hit.

After calculating the brand’s unit economics, we worked out that for one of the products we needed to hit:

1.45 ROAS to breakeven.
1.6 to hit his target ROAS.

As long as we consistently hit a 2.9 ROAS on remarketing, we could scale at a 1.4 ROAS on prospecting to hit the target of 1.6 account-wide.

If we aimed for a 2 ROAS account-wide, we can scale at 1.8 on prospecting as long as we maintained over 3 on remarketing.

Secondly, we analyzed and aggregated the existing creative and audience data across several hundred thousand dollars in ad spend – to identify what was already working in the account and iterating from there.

Alongside our full-funnel account structure, we have our creative testing campaigns. I know we repeat this often, but creative is the most important part of the ad, hence our emphasis and focus on creative.

Overall, at the time of writing this case study, we’ve tested 312 unique creatives, and 36 audiences.

As you can see, we focus more on testing creatives instead of audiences.

With a mass-market offer like skincare, how varied are the audiences going to really be?

Many middle-aged women feel like they experience fine lines and signs of ageing, thus there won’t be too much of a difference in customer psychology if we vary our targeting.

We let Facebook’s algorithm pick out our customers who are most likely to purchase whilst serving them optimal creative.

We were able to test different creatives such as product shots, model shots, user-generated images, videos, product photos with benefit statements etc.

The final step to unlock scale was to eliminate the initial CRO bottlenecks. Our in-house developer and CRO expert Josef worked on improving the site CRO. Bandwidth draining plugins and apps were removed to decrease page load time. Conversion killing bottlenecks such as the site’s poor usability, checkout flow and intrusive pop-ups were fixed to improve site conversion rate and ROAS.

Step 3: Optimize & Scale

Now that we identified and fixed the initial bottlenecks, how did we scale the account to profitability spending $250K/mo and producing $500K/mo in revenue – at 25% higher than the target ROAS?

The biggest lever to pull here was creative. We scaled our winning creatives on high spends whilst continuing to test new ones in the background. Once we found what set of creatives worked, we created a feedback loop with our client, to keep requesting images of the models/product shots that worked. This way, we always were iterating and improving the biggest lever to unlocking scale and improving ROAS.

For example, we started with testing product shots, pictures of models, pictures of ingredients and more etc.

After we find what variation works, we take existing elements of the working creative and continue to iterate and test.

For example, product shots and models worked well, so we would combine them both and run them as a test. We’d then test different models on the same product images, to figure out which model converts best, after we identified that model shots worked.

We tested and iterated a wide variation of creative styles, and eventually boil down the winning elements through our creative testing process.

Another way to further unlock scale out of the account was to improve the retargeting ROAS. This would allow us to scale at a lower prospecting ROAS and get more customers, volume and overall net-profit. We focused on improving the variation of messaging in retargeting such as:

Social proof/review ads – convert those who are skeptical about trying skincare from a new brand.
Reminder + Benefit ads – convert those who were on the fence and just need a second nudge to win them over.

We also tested different angles on our presell pages. For example, writing articles about our facial oils for ageing skin and one for acne-prone skin, and another about our cruelty-free facial oils. Testing different angles on the presell allowed us to find the best converting one to drive maximum revenue.

Improving the initial AOV was another way to improve account ROAS. Instead of selling just one product, we consulted with the brand owner to create bundles to improve AOV and average order quantity (AOQ).

We offered a free travel-sized bottle if the customer bought the full-sized bottle. This incentivized customers to buy the larger bottle, which in turn helped boost our AOV, ROAS & profit.

We also offered a matching body oil with the facial oil, and suitable night serums to complement the day serums.

In this biweekly report, you can see here in the picture how the jump in AOV at the end of March helped us scale:

That way, we offered a tangible benefit to the customer whilst improving our AOV. Winning on bundles and maximizing AOV is a great strategy to improve ROAS, as opposed to relying on discounts – which hurt profit and brand equity.

We also used automated rules to optimize the ads 24/7. Previously, the brand owner was checking the account twice a day. With rules, we automatically checked the account every 15 mins, and automatically turned off ads below our target ROAS. This ensured unprofitable ads and declines in performance were dealt with promptly so ad spend wastage was minimal.

With the right foundations in place, scaling came pretty straight forward. There were no fancy hacks or tactics used here, we just steadily increased budgets whilst keeping an eye on the ROAS. We’ve been able to just keep increasing budgets whilst maintaining a 2X account-wide ROAS.

You can see how during April, we saw initial strong ROAS and doubled down very quickly, spending $15K-$22K/day, and hitting $40K+ revenue days

At this stage, further improvements to CRO, customer LTV, and developing other product lines will allow us to unlock further scale.


In October 2019, our client saw one of these case studies.

And unlike other eCommerce brand owners, who attempt to piece together different bits of content, trying to figure things out themselves, continuing to stay stuck and tolerate the same disappointing results….

He instead took action and secured his Facebook ads audit call with us.

Want the same results for your brand? Then become an action taker today. Book your free Facebook ads audit call below, and we’ll go over the initial analyze phase with you.

We’ll diagnose exactly what the bottlenecks are in your account that are stopping you from scaling to your target ROAS and revenue. It may not be what you think…

Click below to claim your spot in our diary today.