Is having a high customer support ticket volume ever a good thing?

The first question senior leadership usually asks when looking at a support dashboard is whether ticket volume is up or down. If it's up, the instinct is to fix something. If it's down, the instinct is to celebrate.

Neither reaction is quite right. Ticket volume is a number without a story, and if you know anything about data analysis, it’s that the story matters more than the number.

Across 100+ talks at our Customer Support Summit events over the past two years, practitioners from companies handling hundreds of thousands of contacts a month have said the same thing in different ways: volume tells you something happened, but it can't tell you what to do about it until you know why it happened.

The same number can mean your product has a critical bug, your customer base is growing fast, your self-service is broken, or your team is doing its best work. But typically, treating them the same way is where things can go very wrong.

Volume and growth often rise together

At the most basic level, high ticket volume can simply mean you have more customers. And that’s more or less basic math, rather than an out-and-out problem.

Thiago D. Garcia, VP of Customer Experience at VTEX, described his company's position before its customer support transformation:

"Headcount growth in support tracked ticket growth linearly. So the more customers we onboarded, the more tickets they put in, the more people we had to hire."(Customer Support Summit New York 2026)

At a pure growth company, that's to be expected. And yet, the problem wasn’t that volume is high; the real issue was that, in Thiago’s org, ticket volume was scaling with headcount rather than being absorbed through better infrastructure or self-service.

The same dynamic was raised by Simon Rohrbach, Co-Founder & CEO of Plain, in his talk at Customer Support Summit San Francisco 2025. Simon had watched this play out at a food delivery company before moving into B2B support tooling:

"For every incremental 100 orders we were adding per day, we were actually able to add a couple of people in customer support. So we weren't getting more efficient the more we were growing, which was a really scary place to be because it basically means the more successful you get, the closer you get to bankruptcy."

Despite what first impressions might be, ticket volume rising alongside customer growth isn't automatically a failure. The real failure is when it becomes the sole thing you measure and absolutely nothing changes the ratio. (And that’s when it’s sweaty palm time!)

High volume that nobody reads is wasted data

Ticket volume rising alongside customer growth does carry real value, but only if the organization has the infrastructure to extract what it contains.

Tammy Kearns, Group Director of Enterprise Contact Center Products at Walmart, described how her customer support team went from running contact volume linear with growth to actively working upstream with product teams once they'd invested in the right insights infrastructure.

Her framing of volume was candid:

"Quite honestly, talking about contact volume, handle time, return or repeat contact rate to anyone outside of the contact center? That doesn't resonate. It simply doesn't. Is it going to drive gross merchandise value? Is it going to improve customer lifetime value? If you're not connecting in those levels, then it's just a you problem." (Customer Support Summit New York 2026]

The tickets keep coming in thick and fast; they still cost money, but their value to the rest of the business depends entirely on whether support has translated them into language anyone else cares about. A high volume of tickets that nobody analyzes just costs money. The same volume, properly categorized and tracked, tells you exactly where the product is breaking down.

This distinction between generating data and actually using it came through clearly in a separate talk at Customer Support Summit: B2C Edition 2025 from Latha Uttam, Director of Customer Experience at Dell Technologies.

During her talk, Latha discussed how she worked through her team's experience with product changes:

"By tracking repeat support cases and the time spent to resolve those requests, I was able to influence the product team with the help of sales to permanently change the out-of-the-box settings of this feature. Instantly, my call volume related to this was reduced by almost 100%."

That's the payoff. High volume pointed to a broken product setting. Latha’s team documented it, made their case and the problem disappeared from the queue. Refreshingly, a large ticket volume served its purpose in this instance.

Repeat volume is the kind that should worry you

It’s a universal truth that not all volume is equal. The number that genuinely warrants alarm isn't total contact volume; it's repeat contact volume, specifically cases that come back because the first resolution didn't hold, or because the same underlying issue keeps generating tickets.

To no one’s surprise, Latha Uttam hit the nail on the head again; in the same talk at Customer Support Summit: B2C Edition in 2025, she was direct about what systemic repeat volume signals:

"If we solve an issue as a one-off case, then it will keep repeating. We'll be in this vicious cycle of solving problems continuously at a superficial level."

Latha described Dell's approach of using volume trend charts as the first diagnostic step, precisely because the direction of volume tells you something different from the absolute number.

Ticket volume spiking up across a particular issue type means something structural has gone wrong. Ticket volume trending flat with periodic spikes might mean seasonal causes. Ticket volume trending down in a specific category means either the product improved or customers gave up trying.

That's the distinction between volume as a performance problem and volume as a diagnostic resource. If repeat tickets are rising, it's worth asking what root cause hasn't been addressed yet. The tickets are the symptom. The unresolved issue is the cause.

The self-service trap: Volume that looks like a win

One of the more counterintuitive points raised across these conversations is that volume dropping after a self-service update isn't necessarily good news.

Marie Cervantes, Senior Customer Success Manager at Expedia Group, gave a thought-provoking talk at Customer Support Summit: B2C Edition in 2025. In it, Marie described a pattern her team saw repeatedly:

"A well-intentioned help centre update led to a spike in our chat volume. Why? Because the customers did find the article, but they left more confused than before. So we mapped it. Customer lands on the help article. They try the self-service. They still aren't clear. They open the chat with a live agent, but they continue to escalate. It wasn't an agent problem. It wasn't a process problem. It was a design problem."

This is a case where volume going down would have looked like success on a dashboard, but the underlying issue, a confusing help article, would have remained. Customers who didn't escalate would have left frustrated. The ones who did escalate added load to the queue. Funnily enough, neither outcome reflects the kind of contact rate reduction that actually signals product improvement.

Daniel Bunton, former Head of Customer Support at Cleo AI, spoke at length about the difference between contact rate reduction that comes from fixing things for customers versus contact rate reduction that comes from making it harder to get through.

At Customer Support Summit London 2025, Daniel explained a little bit about Cleo’s backstory: it grew to just over a million subscribers, with around 75,000 conversations a month, and he tracked contact rate alongside subscriber growth over five years. The metric he focused on wasn't total volume; it was contact rate per subscriber. When that came down, it meant the product was getting easier to use. When it crept up, it meant something had broken.

"Contact rate is your number one lever. You can really see the correlation over the last five years of reducing contact rate, making your product easier to use, and seeing the benefits of that in customer support."

A five times reduction in CS cost per subscriber. But not from deflecting customers who needed help, but from making the product good enough that fewer people needed to ask.

Customer Support Reimagined with Generative AITransform your support strategy with generative AI. Learn how Crestron boosts CX, efficiency, and agent impact without losing the human touch.Customer Success CollectiveCustomer Success Collective

When volume tells you the AI is working against you

AI deployments have added a specific wrinkle to the ticket volume conversation. Several practitioners described situations where ticket volume increased after an AI rollout, and the reason was the AI itself. Not ideal...

Burak Kebapci, Senior Director of AI Customer Support at Cardlytics, described a direct case he'd experience while giving a talk at Customer Support Summit New York 2026:

"We let our agent handle incidents. But because of a lack of nuance, the agent missed which customer was actually recovered and which wasn't. It sent the wrong update to customers that won't recover. What did that lead to? This ended up losing trust_with the customer and we had ticket volume increase."_

This is a textbook case of ticket volume as a warning sign of a system failing, not a product bug or a service failure in the traditional sense, but a poorly designed AI agent creating confusion at scale. The signal was abundantly clear: the major fix required redesigning the agent's guardrails.

Sarah Sherwood, VP of Customer Success and Support at Typeform, raised a related problem that comes up when leadership interprets volume the wrong way post-AI-deployment:

Sarah described the pressure she faced from leadership after AI deployment. Volume hadn't dropped dramatically because AI had cleared out some of the more simpler queries, what she was left with was far more complex. "It's not bad that the volume isn't dropping," she said. "They're just coming in for different reasons."

At Typeform, AI had deflected the transactional queries, so the remaining volume was more complex. Agents were spending more time per ticket, and offering consultative support where relevant. Handle time was up and volume wasn't dramatically down as they’d hoped. Sarah admitted that both numbers looked wrong on a dashboard, as if they were built for a different era.

In the end, Sarah’s team was doing better work with the customers who most needed it.

AI in customer support: How to build trust before it breaksAI will fail. The question is whether you’re ready for it. A guide to responsible AI deployment for customer support leaders.Customer Success CollectiveOlga Rais

What volume is actually worth measuring

So what's the right way to read volume? Across these conversations, a few consistent approaches emerged.

Contact rate per customer is more useful than total volume , because it strips out the growth effect and tells you something about whether the product or service experience is getting easier or harder to navigate.

Repeat contact rate deserves its own measurement. A customer contacting support twice about the same issue in a short period is a different signal from two separate customers each contacting once. Most dashboards don't separate them by default.

Volume in context of what AI is handling vs. what humans are handling. As Daniel Bunton argued, measuring one aggregate number when the composition of the queue has fundamentally changed gives you the wrong picture.

And perhaps most directly: volume trending up after a deliberate self-service or AI change is worth investigating immediately, because it suggests the change may have created friction rather than reduced it.

The conclusion

Now, none of this means high ticket volume is fine. Support ticket volume that’s high because the product has recurring defects, or because self-service keeps failing, or because the same customers keep coming back with unresolved issues, is a problem. The question is whether you know which category you're in.

The teams that read volume well tend to have invested in how they categorize and tag tickets, how they track volume by issue type over time, and how they translate contact patterns into language the product and engineering teams will act on.

The teams that read it poorly tend to have a single volume number on a dashboard, a target to make it go down, and no agreed-upon explanation for why it moves.

Volume going down can mean the product is improving. It can also mean customers have stopped asking for help because they've stopped trusting you'll give it.

The number won't tell you which one it is.