[Case Study] How We 2x Our Client’s BFCM Revenue to $142k

This client has been in business for 5 years already.

They started out as an Etsy store and eventually grew out of it, building their own store.

Their sales have dwindled over the past 2 years went they came to us.

When they came to us, it was the period where every savvy advertiser should already start preparing for Black Friday Cyber Monday (BFCM) promos.

So we focused our efforts on there.

2018 Results

In 2018, they managed to generate only $68k for their BFCM promos, with $46k attributed to Facebook Ads at a 2.69 Return on Ad Spend (ROAS).

No alt text provided for this image

2019 Results

Here’s what we managed to hit for the same period, taking into account that BFCM 2019 happened 1 week later than 2018’s.

No alt text provided for this image
No alt text provided for this image
No alt text provided for this image

Results were $142k generated, with $63k attributed to Facebook Ads, at a 3.45 ROAS. And we hit 10x ROAS on the first day of the promo.

Goal-setting, KPIs & Budget Planning

It was great to have a clear target to aim for. To do better than what they did back in 2018.

Since the holiday season is an anomaly, and majority eCom businesses make 50% of their annual revenue during this period, we had to use a different calculator for setting KPIs.

The things we took a look at:

  1. How the CPM changed throughout the months for their account in 2018
No alt text provided for this image

2. We then calculated the Delayed Attribution Multiplier that would happen for 2019 based on past data.

No alt text provided for this image

This is an important metric because people would be holding back from making a purchase prior to Black Friday. So the purpose is to calculate what sort of delayed attribution we should be using to account for those people’s behaviors so that we can gauge how successful our campaigns are during this competitive period.

3. Setting Q4 goal + calculate revenue distribution

We then set how much revenue we would like to generate for Q4 2019, and then used revenue distribution data from 2018 to guide our monthly KPIs for Q4 2019.

No alt text provided for this image

4. Calculate revenue generated from Facebook Ads

We then calculated what percentage of the total revenue would be generated from Facebook Ads, so that we can properly allocate the adspend required to hit our goals.

No alt text provided for this image

5. Here’s the final data that is calculated to give us our KPIs for Q4

No alt text provided for this image

With our warplan laid out in front of us, we got to work.

Simplify the Facebook Ads Account

Facebook ads has drastically changed over the years, with machine learning playing a huge part now and moving forward.

We followed Facebook’s Power 5 framework, which consists of:

  • auto-advanced matching
  • account simplification
  • CBO
  • automatic placements
  • dynamic ads
No alt text provided for this image

In the past, it would have made sense to create lots of different campaigns with lots of different adsets to target lots of different people with your creatives.

With Facebook’s machine learning being a core part of the advertising platform, it needs to be fed more data in order to learn the best people to serve your ads to.

As a result of that, Facebook is suggesting the best practice of simplifying your ad account by reducing the amount of campaigns & adsets you have, so that there’s more firepower (aka budget) behind each adset. When there’s more budget being distributed to less adsets, Facebook’s algorithm is able to properly learn the type of audience it should best be targeting for you.

Because of this, for BFCM 2019 we only ran 15 different campaigns with only 4-5 adsets within each.

Our campaigns can be segmented by

  • warm + hot audiences
  • cold audiences
  • DPA campaigns
  • different geographical campaigns

In the past, we would have used 18 different retargeting audiences that have worked really well for us. But with the simplification framework, we grouped them up to only 4 distinct audiences:

  • People who have visited the site
  • People who have engaged with Facebook page or Instagram profile
  • People who have viewed 50% of our videos
  • People on the email list

The warm campaign did extremely well, achieving a 29x ROAS on the first day of the promo.

Having only 4 distinct audiences to manage for each warm campaign meant lesser time spent on management, which allowed us to focus more on strategy and messaging.

Prepare lots of creatives with different angles, suited for different days

Which brings us to the next reason on why we got great results for this client. Since our time spent on managing the campaigns have been drastically cut down by following Facebook’s simplification framework, we had more mental bandwidth & time to focus on creatives.

Since our BFCM promo was 13 days long, we would risk our audiences getting fatigue with seeing the same creatives. So we had to come up with different angles suited for different audiences based on the different days.

We had slightly different messages for the days that our promo would cover, which includes:

  • the day we launched our promo
  • the day before Thanksgiving
  • Thanksgiving day
  • Black Friday
  • the day after Black Friday
  • 24 hour & 12 hour countdown for our Black Friday promo
  • the day before Cyber Monday
  • Cyber Monday
  • 24 hour & 12 hour countdown for our Cyber Monday promo

As you can see, when we split them out like this, we can tailor slightly different messages, and reduce the chances of getting ad fatigue & burning our our audiences.

Use CBO for maximum ROAS

Another thing that’s part of Facebook’s Power 5 is the usage of Campaign Budget Optimization.

When CBO first came out, people were getting mixed results with it. But now that it’s very powerful, we wouldn’t launch any campaigns without using CBO. Unless there’s a reason we need to use the regular ABO (adset budget optimization) for testing purposes.

If we used ABO for our highest performing campaign, I doubt we would get anywhere near the 24x ROAS we managed on the first day.

Don’t be afraid to use Broad audiences, split by countries

In the past, prior to utilizing Facebook’s machine learning, the best way was to specifiy who you want to target. But with the Power 5, you would be limiting Facebook’s ability to find you the best person that would convert.

As a result of that, huge audiences work better these days compared to the super-tightly defined audience.

Since this client has been running ads for at least 3 years, they have lots of data on their customers in their Facebook ad account.

When you have lots of data, that allows Facebook to give you great results when you use broad audiences. And by broad audiences, I mean keeping the targeting to all ages of 18-65+, all genders.

But it’s best to split them up by countries, because if you don’t, initially the budget will be allocated to the countries with the biggest sizes, and that could mean 3rd world countries that aren’t able to afford your products at all.

For this campaign, we split the broad audiences like this:

  • US, UK, Canada, Australia
  • Europe
  • Rich Asia
  • Scandinavia
  • Worldwide (excluding the ones we’ve already targeted above and also bad countries that, based on data, we know will never ever purchase)

Warm DPAs help scoop up remaining sales

No alt text provided for this image

Warm DPAs (Dynamic Product Ads) are amazing. They’re so simple to set up and their returns are super high. We were getting 4.33x ROAS throughout the whole promo.

Since we were targeting broad audiences for our cold campaigns (meaning we’re targeting people who’ve never come across the brand before), the ROAS would no doubt be the lowest amongst the other campaigns.

So these audiences would definitely need additional touches in order to convert.

Here’s where DPAs are great, because they are only shown to people who have browsed at least 1 product. And the ads they get are basically the products they’ve browsed but haven’t purchased yet.

This serves as a reminder for them to take a look again at the product they showed interest in.

DPA campaigns also had the highest frequency amongst the other campaigns, at 7x. So basically our DPA ads were shown on average 7 times per person. The campaign with the next highest frequency was our warm campaign at 4x.

Cold campaigns should at least be at breakeven ROAS at scale

When you’re playing at the level where your daily adspend is 4-5 figures, no doubt your cost per purchase will increase, resulting in a lower ROAS.

Instead of thinking about how much lower your ROAS is compared to before, you need to rethink the way you look at your whole campaign.

Cold campaigns are just a way to fuel your warm campaigns.

And warm campaigns have a drastically higher ROAS.

In order to sustain your warm campaigns, there needs to always be new people coming through the cold campaigns.

In order to make the economics work, the cold campaign therefore needs to be at least breakeven.

When we scaled to $2k per day for this client, ROAS at the cold campaign level dropped, which was normal, but that also allowed us to increase the budget for the warm campaigns, which sustained its ROAS.

Which meant more revenue and more profits for the client! Ka-ching!

Clarity trumps persuasion

Sometimes you see big brands having complex offers during a sale. For example:

  • 10% off the first $50 purchase
  • 15% off the next $100 purchase
  • $100 off purchases with 3 orders and above, then 20% additional for 5 orders and above, with a free gift for purchases above $500

While that might seem smart from the brand’s perspective, which does seem to reward the consumer for spending more… But it’s more complex than a straight up, 25% off sitewide offer.

And when consumers have to do a lot of math to figure out how good the offer really is, they get frustrated and would end up not making a purchase.

As a result of that, we advised the client to do a simple 40% off sitewide offer. Site visitors wouldn’t have to spend time figuring out if the item they wanted was eligible for the offer, they didn’t need to do mental math.

It was straightforward and simple.

And if sales performance was low, we didn’t have to deal with too many variables to figure out why exactly the sale didn’t yield the results we were looking for.

Evolve your email marketing

Email marketing is still king. 60% of our revenue came from our email list.

The best thing about email marketing is that you own the list, you don’t need to pay a dime to Facebook for sales generated thru email.

For the BFCM promo, we sent 7 different email campaigns over the 13-day period.

And each campaign is resent to people who didn’t open 24 hours after the initial email.

We also excluded the people who have purchased during the BFCM promo from receiving any more campaigns.

This step is important because you don’t want to burn out your list.

The way to do that is by integrating your email marketing platform with your ecommerce platform so that they are able to share data with each other.

Doing this allows your list to be super segmented so that your emails can be super targeted.

Apart from that, your Facebook ad campaigns can also be more powerful because it’s fueled by fresh data coming from your lists & the proper segmentation being done in there.

Use Aftership for a smooth post-purchase customer experience

We integrated Aftership with this client’s Shopify so that our email marketing & ad campaigns can be more accurate.

Aftership is a shipment tracking solution, so that you’re able to know when each individual customer has received their order.

Without this solution, you’d be flying blind in terms of knowing when your customer has received your product.

Sure, you could estimate the shipping time & use that to target your email & ad campaigns based on that. But wouldn’t it be more powerful if your customer receives the right message at the right moment?

Here’s what you can do when you know when exactly a customer has received their product (these are best implemented using an omni-channel strategy, meaning using email marketing + Facebook ads):

  • get their review
  • 2 days after (or however many days it takes for your customer to fully experience your product), ask them to recommend your product to their family & friends in exchange for some form of incentive (gift card, coupon code, etc.)
  • 7 days after, cross-sell them other products

Eventhough this step did not directly contribute to the $142k BFCM promo, it does improve the customer experience, which will result in higher satisfaction in the customer, and lead to more word of mouth & better reviews.

So there you go, the few important steps we took to 2x the revenue generated for this client and increased 28% higher efficiency of their adspend usage.

If you’re looking to take your eCommerce brand to the next level, reach out to us so we can give you an audit & see what percentage of improvements we think we’re able to get for you.