Our own mask store scaled from 0 to $1,606,669 USD at 220% ROAS in 56 days using Taboola & Facebook Ads while donating 20% of profits to local elderly care centers.

We were getting a lot of requests from businesses selling masks asking to work with us, but we never felt any of them were a good fit.

Either they didn’t have good quality products or their supply lines weren’t taken care of. Some of them didn’t have legitimate certificates or we just didn’t feel the owners were ethical or willing to do some good.

This is why, when our logistics partner suggested to us  the idea of an ethical mask business, we jumped on board.

We had a good ride, but when our CPAs started to increase bid time (despite all our efforts), we decided the hassle was not worth the profits anymore and we closed the store.

This is why I’m able to share the full store and campaign details

We created 1 worldwide store, 1 store specific for the UK market and 1 store for the US tests. We also had more backup country-specific stores with parts of USP specific to those countries ready to deploy, in case we saw strong results from a particular country.

Our stores were one-page, one-product stores with no collection pages and no cart page.

We implemented all our best CRO practices and our conversion rate jumped to an astounding 4%+. This is when we knew we had a winning product and a winning funnel.

We used https://brandfeatured.com/  (400USD service) to get brand mentions to boost social proof

Infographic renders & 3d videos ordered from Fiverr for $50 or so

Stand by expert recommendations and trending industry news to borrow their social proof

We quickly got reviews from friends, team members and imported a lot from the manufacturer to get us going for the first week or so (before the first customers received the product). We also gave out free products in order to accumulate lots of reviews—fast!

Then, we collected reviews from happy customers using our customer service team and by offering a deep discount on the next order via an email sequence.

We also used Fomo – Social Proof Marketing app to boost social proof and trigger first-time purchases. There’s mixed evidence about the effectiveness of this tool, but we used it anyway.

Implement tiered discounts to boost AOV. We did not use any post-purchase upsells, as we were concerned about the revenue tracking issues with Zipify.

Emphasize the deal by comparing current price to a “regular” price

Test pricing in a controlled environment, e.g. test for a limited time on a single market.
(Warning: Taboola’s algorithm really doesn’t like changes on the landing page and pricing changes can disrupt performance of stable campaigns.)

• Always have multiple payment options. In the most common markets (US, Canada, EU, Australia, New Zealand and other ePacket countries), you’ll capture most benefits by using Shopify payments, Paypal and Google Pay.
• Many payment processors will introduce temporary payment holds on a % of your sales. Prepare for this, and remember it for when you’re making cash flow projections to not run out of money when scaling ads.

Use badges below the Buy now button that emphasise parts of your USP, especially delivery times & safe payment methods.

• Set up a worldwide store and worldwide Facebook ads campaigns to test countries to expand into\
• Then, once you know which countries performed well, consider setting up country-specific stores with translated copy for larger markets.
• Use currency converter (in this case, we used Auto Currency Switcher app)
• Most EU countries don’t need translated copies to perform well (especially Netherlands, Sweden, Norway, Iceland, Finland, Denmark).
• Some countries, such as Brazil or France, are unlikely to perform well in the early tests unless your store is translated and you’re using local payment processors.

• We set up basic cart abandonment and post-purchase automations and we were sending weekly email campaigns to squeeze more results.

• We used SMS bump for cart abandonment retargeting and post-purchase upsell sequences.

• We placed brochures with discount codes on repurchases in the packages to increase LTV.

Even if you have the best product and excellent reviews, when you scale, you might struggle with Facebook Customer Feedback Score. Here are a few reasons why and how to deal with them.

• Competitors maliciously attacking your feedback score
  • Exclude from your ad sets: Behaviours: Facebook Page admins or Business Page admins
• Shipping time:
  • Invest in a strong customer support team to answer customer questions within a few hours
  • Set up frequent (1-2 times a week) email communication with updates on the product shipping to show your customers you haven’t forgotten about them
  • Set your shipping time on your Facebook Customer Score Page (https://www.facebook.com/ads/customer_feedback/)  to 9+ weeks (this is the time when Facebook will send a feedback survey to your customers)
  • Set up order tracking tab on your website
• Product Quality:
  • Gather feedback from customers, work with your manufacturer to adjust the product based on the feedback or change the manufacturer altogether.
• Providers that claim they can fix your score:
  • There is no substitute for improving your product, fulfilment speed and proactively communicating with your customers. Some providers out there claim they can fix your score by deploying a number of agents to leave positive feedback on your ads. These solutions might work at small scale, but if you’re scaling and doing big volume (>1000 orders per day), the number of agents you’d need to keep up with decreasing feedback would be in the HUNDREDS. If you’re dropshipping and doing a volume like that, consider hiring a logistics partner that can help you build a local fulfilment centre network in the countries you’re selling to (remember to consider VAT obligations before making the final decisions).

• We first run a small test to see if our ads would get approved. Once we had confirmation that masks advertising was allowed, we launched on a full scale.
• Accounts
  • We used 4 different ad accounts connected to separate business managers and payment methods to minimise the risk of interruption (in case of account bans).
  • 2 of the accounts got banned in the first 2 weeks.
  • We were then very careful not to experiment with creatives too much. Ad approval and account bans were the main reasons why Facebook was not our leading channel on this project.
  • We didn’t manage to get our business Manager whitelisted, as we did not have enough ad history on this brand.
• We were driving traffic directly to the store pages.
• We used only CBO campaigns with auto placements.
• Countries:
  • We started by targeting UK.
  • Then we launched worldwide and EU-wide campaigns to test countries for expansion. For example, nordic countries were performing very well (Denmark, Sweden, Norway, Iceland). We decided to keep them in the worldwide campaign for stability.
• Targeting:
• We used only broad targeting (all ages).
• Creatives:
  • We used mostly videos. For some reason, images were disapproved every time we tried.
  • We had to stay away from using “mask” wording or any reference to “antibacterial” or references to “WHO” to avoid getting flagged by the automated ad review algorithms.
• Bidding:
  • After initial success with auto bidding, we started scaling aggressively, doubling the budget every day while using Target Cost bidding to prevent overspending.
  • We started by setting the at around 1.5x AOV and, as we saw good delivery, we lowered the bid gradually to £19, at which point it stabilised. Any lower, and our reach was limited—but any higher, our ROAS dropped.

• For an introduction to Taboola ads, check out my video: https://youtu.be/197L9MBtjy4
• We spent a total of £377,757 at CPA of £13.2 (ROAS Tracking was messed up for most of the advertising period, but taking into account our aggregated AOV of £29.32, our estimated ROAS was 2.22)

• Our offer is simple, so we were driving traffic directly to product pages. However, for most offers, I’d use advertorials instead. We’re currently working on a few to test here as well.
• The approval process was fairly straightforward.heir reps are waaaay better than Fb reps and they walk you through the whole process.
• Tracking setup is dead easy, and the reps can help with that too.
• Taboola’s algorithm is super advanced. As you can see in the graph, whenever we bump the budget significantly, CPA goes up on the same day but then stabilises again the following day.
• Creative-wise running more on a news angle; click-baity headlines and images that just break the pattern and make people curious is the way.
• Taboola has a massive reach. It’s most likely even bigger than Facebook, so it’s definitely a great way to scale eCommerce.
• Our campaigns were split by Desktop/Mobile and by different countries we targeted.
• Out top countries:
  • UK
  • Australia
  • Ireland
• Our Top creatives were pretty basic, yet effective. We tested almost a 100 different combinations of creatives and those were still our winners.

• Our CPA, however, was increasing steadily—probably indicating market saturation and increasing competition.
• At some point, our bank (HSBC) blocked our card, which disturbed the algorithm big time. It recovered a bit after we decreased the budgets, but it started to rise again, despite us testing new creatives and countries all the time.
• We made one last desperate attempt to increase the prices to cope with the rising CPA, but it only disturbed the algorithm even more and killed our conversion rates. It only went downhill from there and we decided to kill this channel. In fact, we should have done that even earlier because it would have saved us some losses.

• eCommerce manager – overseeing the whole marketing side
• 1 Logistics manager – managing 1 main factory and 1 backup factory as well as 2 fulfillment centres: 1 in China, one in the UK
• Customer Support team
• Facebook Ad buyer
• Taboola ad buyer
• 2 Creative guys
• 1 Copywriter / Email + SMS expert
• 1 Analytics expert

• We had 4 full-time staff on customer support and our eCommerce manager was helping them (in case of escalations). Total cost: $5k per month.
• Even after we closed the store, we needed to maintain the customer support for the next 3 months (in case cancelations or questions came up).

• We were aiming to keep product + shipping + operating costs below 33% of the revenue (mostly achieved) and marketing costs around 33% of the revenue (we didn’t manage to maintain it, apart from at the start).
• Shrinking profits and high logistics complexity led to our decision to close the store at the end.
• Fortunately, we got a VAT exemption (VAT would eat into most of our profits).

• Partner up with a long-standing fashion brand to get whitelisted Facebook ad accounts.
• Use prepaid cards (e.g. Revolut) or US credit cards for ad spend to prevent interuptions in payments, which mess with the ad platform algorithms and kill ROAS.
• Enter a less regulated market, where we could leverage our marketing more.
• Kill ads sooner when CPA goes up.

We donated 50,000 masks to local elderly care homes and essential workers in the UK, US, Ireland and Australia. That amounted to almost 20% of our profits.