How Thrive Causemetics Turned Customer Service Into a Company-Wide Intelligence System

Brendan is the COO and CTO of Thrive Causemetics, beauty’s largest give-back brand. He’s been there since day one, when he and founder Carissa personally answered every customer email. This is the story of what it took to scale that original commitment without losing it.

The problem wasn’t volume. It was inconsistency.

Thrive’s customer service philosophy has always been simple: take care of them, even if you can’t fully take care of them. That means showing up for every customer, including the person writing in because a family member is going through breast cancer and received products through a charity donation.

“We feel like it’s our obligation to take care of that person in whatever way we can. That still is instilled today.”

That ethos is easy to hold when there are two of you. It gets harder when ticket volume scales into the tens of thousands and customers who reach out twice start getting different answers each time.

Thrive was running Zendesk for most support, a separate tool for social, and a knowledge base that was always in flux. During peak periods, the consistency that defined their brand quietly eroded.

“Especially during peak times, it was inconsistent in the answers we were giving and the timing. We needed to make sure we were reliable to those customers, because they trust us.”

— Brendan

Reliability is a brand promise. Breaking it, even subtly, erodes a decade of trust.

Three non-negotiables, and none of them were automation rate.

When Brendan evaluated AI solutions, he came in with hard requirements.

Brand control came first. A generic AI response, technically correct but tonally off, would do more damage than a slow human one. Second was learning: the system had to adapt in real time and surface gaps back to the team. Third was integration depth. Thrive runs deep in the Shopify ecosystem with Yotpo, Recharge, and more. An AI that couldn’t reach into those systems would create a new problem while solving the old one.

“If you can go into those individual applications and make changes too, suddenly it’s really, really helpful for us to surface that information.”

Siena checked all three.

What changed, and what nobody anticipated.

The operational wins came fast. SLA performance went from something the team was chasing to something they were consistently beating. Peak volatility leveled out. No more emergency staffing calls after a big sales weekend.

But the more meaningful shift happened inside the team. Thrive employs makeup artists and estheticians, people with real expertise in the products and the industry. Before Siena, they spent most of their day on order status and return policies. After, they were finally doing the work they were hired for.

“Suddenly they are more often answering makeup artist and esthetician questions. They can be better stewards of the questions they’re answering.”

— Brendan

Then came the outcome no one fully anticipated. Siena Intelligence now reads every incoming ticket, synthesizes patterns, and sends automated reports directly into Thrive’s Slack. Product development gets early-detection alerts when ticket velocity on an issue spikes. E-commerce queries it for website friction signals. Marketing uses it to pressure-test product hypotheses.

CX data that used to live in a ticketing system is now informing product decisions, marketing strategy, and e-commerce roadmaps. The team didn’t just get faster. They became an intelligence function.

“We have so many tickets, it’s almost impossible to keep up. What’s the real pulse of what’s going on week after week? That’s where it helps us.”

— Brendan

On AI, honestly.

Brendan’s distinction between brands using AI and brands being changed by it is sharper than most. At Thrive, he’s watched people cross team lines they never would have before: finance getting deep into product development, e-commerce collaborating with creative, because they could finally see problems and had the tools to solve them.

“There isn’t your line of what your job role is anymore.”

On fragmented stacks, he’s equally direct. The real cost rarely shows up on the invoice. It’s maintaining multiple versions of your brand’s AI brain across systems that don’t talk to each other. Every disconnected tool is a place where context gets lost.

“You have to have core systems you’re probably not going to move away from, and really invest in those.”

What he’d tell another operator.

“The AI solution is maturing quickly enough to be a suitable solution even at high volumes. We’ve honestly struggled with more enterprise-type solutions, moving slower, trying to bolt on AI. There’s a large difference between providers working at the old pace and the ones rebuilding themselves.”

— Brendan

His advice is practical: enter with a hypothesis, define what you expect at 90 and 180 days, and measure directionally. Not just deflection rate, but time saved, customer retention, and whether your team has finally found space to think beyond steady state.

“Anytime a team gets into that dreaming kind of thing, asking what the best state looks like rather than just managing the inbox, that’s always a good sign for us.”

— Brendan

At Thrive, customer service started as two people who refused to let anyone go unanswered. A decade later, the obligation is exactly the same. The infrastructure is just finally built to match it.

Thrive Causemetics has been a Siena customer and design partner since [year]. They run Siena across customer support, Siena Intelligence, and company-wide CX analytics, all on Shopify.