Cannon Balm

Barely any data to start with, then a platform update switched the tracking off mid-flight.

+33.5%

Revenue

+95.9%

Returning Customer Revenue

+19.3%

Average Order Value

Ad Platforms Used:

About The Brand

Cannon Balm makes an organic pain-relief balm and sells it direct on Shopify. The buyer skews older and buys for a specific ache, which means the product does the retention work if the first order ever happens. Rohan, the founder, is the face of it.

Rohan booked a strategy session with us in October 2024, a few weeks after he'd started running Facebook ads himself.

A few weeks of data doesn't give an ad account much to learn from, so in practice this was close to a cold start.

The Challenge

A near-cold start on a small budget is mostly a patience problem. You're paying for data before you're paying for orders, and acquisition sat near breakeven for the first stretch because that's what a cold start looks like.

Then in January a platform policy change switched off the standard conversion event overnight. In reality that's the part nobody plans for. We moved optimisation onto a substitute event and kept the account trading, but the reported numbers went dark for a few weeks and early-year efficiency took the hit.

Video & Static Ad Creatives

1

Meta Ads

  • Two prospecting audiences, one message.
  • Broad let Meta find buyers on its own, while a physical fitness audience went after people who push their bodies and live with the aches that come with it. Rohan's founder story worked for both, which told us the message was doing more work than the targeting.
  • Retargeting went to everyone who'd visited in the last 180 days, at a cost per acquisition well below the account average. An older buyer weighing up pain relief takes time to decide, so a long window catches people still thinking it over.
  • Cost Cap went in because in health you're bidding against pharmaceutical brands with far deeper pockets. A cap stops Meta paying more for an order than it's worth.
  • When the purchase event stopped reporting, optimisation moved to a substitute event so the account kept learning.
2

Creative

  • We wrote the scripts. Rohan filmed them.
  • Founder-led video suits a product like this because the buyer is trusting a real person with their pain, and Rohan already had that credibility on camera. So rather than hiring talent, the job was writing scripts he could deliver in his own voice.
  • In April we launched 'AI Voiceover on FB Comment Screenshot': a real Facebook comment about the balm, read aloud over a screenshot of it. People know exactly what a genuine comment looks like, so it lands as proof rather than advertising.
  • New edits of UGC from Harry, a creator, went live the same month, and both creatives quickly became some of the strongest in the account.
  • While purchases weren't reporting, creative was judged on the substitute event instead.
3

Offer and Average Order Value

  • The fastest win wasn't in the ad account.
  • We set up bundles on the website in March and a cart upsell in April, and average order value rose 14% month on month without any extra ad spend. When acquisition runs close to breakeven, lifting what each order is worth does more for the numbers than bidding harder ever will.
  • Add a jar to the cart and a pop-up offers 20% off a second one. That works out to only 10% off the whole order, a far cheaper way to lift order value than a sitewide discount.
  • Our advice to Rohan was to email existing customers, because a balm that works gets bought again once the jar runs out. Repeat revenue nearly doubled between February and April.

In Their Words

“I was attracted to the fact that you're putting yourself out there, that you're getting paid based on your results. When people do a good job, they deserve to be paid based on their performance. It was a completely different experience for us.”

- Rohan Cannon, Founder, Cannon Balm

The Results

Across the two-month recovery window from February 2025, revenue grew 33.5%, returning customer revenue grew 95.9%, and average order value lifted 19.3%.

To put that in perspective, this is a small account recovering from a tracking outage, not a scale story. The returning customer number is the one worth watching. It says the product does its job once it's in someone's hands.

There's still work to do on acquisition. That's the honest position.