How Spanx Turned CX Into a Business Driver and Cut Handle Time in Half

We sat down with Kylah, Spanx's VP of Global Customer Experience and Consumer Insights, to talk about what AI has actually changed. Not just for her team's efficiency, but for how she thinks about the CX function entirely.

Kylah has been at Spanx for three and a half years, and recently expanded her scope to include wholesale optimization. She's been working with Siena since late 2023. What she described isn't a story about automation. It's a story about a function finding its seat at the table.

Here's what she shared.

CX three years ago is not CX today

"The CX that folks knew, call it three years ago, is not the CX that we have now. CX has gone from a place that's kind of just answering questions, solving customer problems, to really an organization that's a business driver, responsible for results, responsible for generating revenue."

That shift didn't happen on its own. For Kylah, AI was the accelerant. It gave her team immediate visibility into who the Spanx customer is, what they want, and where they are in the journey — including on a product page, where Siena can step in and help close a sale in real time.

It also eliminated the long lag time that used to exist between noticing something and knowing something. Surveys. Focus groups. Weeks of waiting. Now that data is accessible immediately.

The hiring change nobody's making yet

One of the clearest signals that CX has changed is where Kylah is putting headcount.

"A lot of my peers think: you onboard a tool like Siena and then it just takes care of all your problems. While Siena is awesome, I think to get the most out of it, you need a person completely dedicated to it."

So that's exactly what she did. She hired someone whose sole job is to own the AI agent end-to-end: living inside the product, understanding new releases, QAing responses, and fine-tuning tone and voice.

"Three years ago, I wouldn't have looked for a person and said, 'You're going to sit on the CX team and your only job is going to be to focus on our AI agent.'"

She believes this is where CX hiring is headed. Not more agents to handle volume, but AI strategists — people skilled enough to manage a portfolio of AI tools and make them work together.

Why brand voice and tone won the vendor decision

When Spanx went to market for an AI solution, they talked to a lot of vendors. Good technology. Good CSMs. Plenty of options.

But Kylah had a non-negotiable that narrowed the field fast.

Spanx customers aren't usually coming in with simple questions. They're coming in at moments that matter: a wedding, a postpartum body, a milestone event. They're sharing personal, sometimes vulnerable details about themselves, and they want to feel heard.

"You don't want them to be engaging with an AI tool that's impersonal."

For Kylah, that meant brand voice and tone had to come first. Full stop. She could work backward from there on everything else.

"Siena really felt like a girlfriend, a trusted person that I'd be comfortable sharing personal information with. And when it didn't, the ease of being able to tweak the voice, tweak the tone, tweak the disposition so that it felt that way was seamless."

The outcome nobody expected: AI making human agents better

Kylah expected efficiency gains. She expected Siena to handle volume and free up her team's time. What she didn't expect was what happened to her human agents.

As Siena's responses became part of the team's daily environment, agents started emulating them. The thoughtfulness. The personalization. The way a product recommendation was framed.

"As we implemented Siena, I noticed the responses and the thoughtfulness and the personalization tied to our human agents also became better, because they had other examples of what good work might look like through the responses Siena gave us."

CSAT went up across the board. Not just from AI handling more volume, but from humans performing at a higher level too.

Kylah later heard a similar story from another brand on Siena. Their VP was using the AI's brand persona to actively train her human team. It's an outcome that doesn't show up in most AI ROI conversations, but Kylah thinks it should.

Customers stopped treating it like a bot

Siena now handles 50% of Spanx's conversations. But the number that sticks with Kylah isn't the automation rate. It's the behavior she started seeing from customers.

They were asking for the AI agent by name.

Spanx discloses that it's AI. Customers know. And somewhere along the way, they stopped caring. Or rather, they started caring about the relationship more than the label.

"I expected it to be efficient. I expected it to be good enough. I didn't expect the connection that customers would have to it."

The team isn't in the inbox anymore

When Siena took over tier-one volume, something opened up for Kylah's human team. Time they used to spend heads-down in the inbox got redirected toward work that required actual judgment.

VIP birthday surprises. Complex, high-touch interactions. Surprise-and-delight moments that would have been impossible before.

But the bigger shift was in how the team engages with data. Agents who used to live in the queue are now surfacing insights across the organization.

One example: Siena flagged an unusual pattern around shipping complaints in a specific region. Taken alone, the data point wasn't big enough to register. But the consistency of the signal — customers in that region all saying the same thing — prompted Kylah's team to dig deeper. They found a carrier issue affecting nearly 100% of deliveries to that area. They never would have caught it otherwise.

"The data point wasn't big enough that we would have seen it, but Siena spotted it and noticed right away that something looks a little fishy."

Now the team is surfacing product defect signals, pulling positive customer quotes for the brand team to use in marketing, and feeding product teams real-time feedback on what customers actually want. CX is no longer siloed. It's a feed that every department is starting to tap.

The metrics moved, and so did the boardroom conversation

Kylah tracks the usual stack: CSAT, NPS, average handle time, SLA. Since implementing Siena, average handle time has been cut in half. CSAT is up five to seven points.

But what the numbers did beyond the CX org matters just as much.

"It's been easier for me to make the case for the customer to our ELT, to our board, because the data is right there. It's easier to illustrate the story and tie metrics and revenue numbers to customer experience in a way that I could do before, but it just took a lot more to do it."

That's the real unlock. Not just improving the numbers, but finally having a language that the rest of the company responds to.

What's next

Kylah's team is now working on a shopping agent: an AI-powered concierge available on-site that can help customers navigate products, answer questions, and move toward a purchase based on their shopping behavior, return patterns, and preferences.

"The ability to connect the dots throughout the experience — they might recognize who you are and recommend products they know you like. We're excited to see what potential revenue comes from that."

She sees the shopping agent as the next chapter in making CX a provable revenue channel. Not just defensible, but demonstrable.

Her advice for skeptical brands

For brands still on the fence, Kylah's guidance is straightforward.

Start slow. Pick one or two pain points. Get it right before you expand. And don't confuse implementation with completion.

"A lot of companies think: I'm going to partner with this tool and implement it and that's it, all will be well. It's really worthwhile to hire a person, maybe two, whose sole job is to think about AI for CX."

She uses a tennis analogy. The first time you pick up a racket, you're not going to hit the ball cleanly. But the more time you put in — refining, tweaking, working through the bad swings — the better it gets. You're probably not going to see everything you need in the first 30 days. Maybe not even 60 or 90.

But when you do the work, it pays dividends.

"You have to put the time in. And when you do, it really pays dividends.