Inside 100% Pure’s “risky” experiment that showed email beat Facebook on retargeting
In August 2022, clean beauty brand 100% Pure stopped all Facebook advertising.
It was a “really risky” experiment, says Ric Kostick, CEO and co-founder of 100% Pure. Since its inception in 2005, the primarily DTC retailer had relied heavily on Facebook throughout its customer acquisition funnel.
Kostick and his team used Facebook for top-of-funnel awareness advertising, and for retargeting visitors to their online store who hadn’t yet purchased.
But business as usual wasn’t working.
July 2022 was an unexpected disappointment for 100% Pure. It had been a lucrative month historically, but the company didn’t turn a profit—even though every channel hit its return on ad spend (ROAS) goals.
Kostick had a hunch that 100% Pure had redundant spend in their funnel—specifically, that they were overspending on Facebook to drive purchases that email would have driven anyway.
He had to take a risk to find out if he was right.
- How the profitability problem escalated
- A month without Facebook ads
- The new and improved marketing strategy
How the profitability problem escalated
Customer acquisition doesn’t go from profitable to unprofitable overnight. In hindsight, Kostick thinks the issues that flared in summer 2022 began building in 2019.
Two main things destabilized 100% Pure’s marketing measurement: KPIs and attribution.
The issue with ROAS
In 2019, 100% Pure started outsourcing more marketing work to agencies—usually 1-2 at a time, each focused on a specific marketing channel. The agencies set ROAS goals. And hit them.
It looked like a success story. But Kostick knew it was risky to use ROAS as a primary marketing KPI.
“I’m an oddball because I’m anti-ROAS,” Kostick says. “You can easily manipulate ROAS to make it seem good.”
It’s an inherently optimistic metric. ROAS doesn’t factor in non-marketing costs involved in connecting a customer with their purchase—whether that’s per-unit costs, like production and shipping, or fixed costs, like office space.
Still, back then, Kostick didn’t think it was worth fighting. The business was humming along smoothly. The agencies did strong work. ROAS reigned.
The decline of last-click attribution
Then, in 2021, iOS 14.5 rolled out. Facebook ad attribution as everyone knew it disappeared.
“We knew it was going to be challenging,” Kostick says. “It did strike me—strategy’s got to change when this comes out.”
Before iOS 14.5, the 100% Pure marketing team had relied on Facebook’s last-click attribution.
Last-click is not a perfect marketing attribution method, as any seasoned marketer will tell you—it undervalues top-of-funnel channels. But it has one thing going for it: simplicity.
There’s one last click for each purchase. Use last-click attribution across multiple channels, and the revenue each channel claims to have produced will reliably add up to total revenue.
During the iOS 14.5 rollout, Facebook Ads stopped supporting last-click attribution and shifted to attribution windows—so purchases made a certain number of days after a user clicked or viewed a Facebook ad got attributed to Facebook.
But of course, a user might see a Facebook ad, an email, and a Google ad all in the same day.
So what happens when more than one channel can take credit for a single purchase? Chaos. Double-counting. Sky-high ROAS, despite middling revenue.
“That created the playing field for a lot of overlap and over-inflating,” Kostick says.
The final straw
In summer 2022, Kostick’s team hired a Facebook agency and set them a ROAS target—which they met.
But when Kostick added up the revenue all of 100% Pure’s marketing channels claimed in their ROAS calculations, it was 3x the amount 100% Pure had actually made.
And it wasn’t a profitable month for the company.
ROAS had always been an imperfect metric, and the attribution changes around iOS 14.5 made it an even less reliable KPI—to the point where it was now misleading.
“I realized, this is all messed up,” Kostick says. “We need to change our model completely because of this iOS update. We can’t rely on what everyone’s reporting.”
A month without Facebook Ads
In August 2022, Kostick and his team stopped Facebook spend, which left them with gaps at the top and bottom of their funnel.
At the top of the funnel, they relied on cheaper awareness plays, like broad Google Ads campaigns.
For retargeting, they relied on email marketing. This meant they could only retarget shoppers who signed up for emails while shopping—whereas with Facebook Ads, they could retarget nearly anyone on the planet.
Would it work?
In the end, it did. “We went from negative to positive during a tough time, coming out of the impact of COVID,” Kostick says.
August results looked like this:
- Margins: huge MoM improvement
- Revenue: slightly below goal
Especially for retargeting, Kostick’s hunch had been right.
Facebook could retarget a wider array of shoppers, but the ones who converted typically subscribed to email, he explains.
The new and improved marketing strategy
Today, Facebook retargeting is over for 100% Pure.
There was too much overlap between paid social and email retargeting, which the brand runs through Klaviyo. “Moving forward, Klaviyo is going to be the core of our bottom-of-funnel,” Kostick says.
The marketing team has strengthened their retargeting with digital fingerprinting. Instead of cookies, this tracking technique relies on each user’s unique digital fingerprint—made up of publicly available parameters like IP address, screen width, and device type.
This helps 100% Pure understand when and how email subscribers visit their site, even when they’re on iPhones.
Kostick’s team still spends on Facebook, but only for top-of-funnel campaigns. The new KPIs for these? “Old-school” metrics like cost per click (CPC) and cost per thousand impressions (CPM), Kostick says.
No more ROAS.
Looking back, the first years of the 2020s were tough to navigate, Kostick notes. The combination of COVID and iOS 14.5 broke all established best practices in ecommerce.
Now, growth at all costs is out. Margins are in.
“A lot of businesses during this time, they got smaller—like we got smaller—but they got more profitable,” Kostick says. “We have a good solid strategy moving forward for customer acquisition and converting potential customers into fans.”