UJET vs Five9 and the Fight for the AI Contact Center

The contact center is the software category nobody brags about at dinner. It runs the phone queue, the chat window, the "your call is important to us" hold music. For twenty years the pitch barely changed: move the call center off old phone lines and into the cloud. Then customer service turned into one of the clearest places where AI actually earns money, and two companies that could not be more different in size started chasing the same prize.

On one side is Five9, a public company that has been at this since 2001, now doing more than a billion dollars a year and growing its AI business faster than anything else it sells. On the other is UJET, a smaller and quieter company founded in 2015 that most people outside the industry have never heard of - which is strange, because a lot of people have used it without knowing. When you contact certain companies through Google's own contact center product, UJET is the machinery underneath.

Put them next to each other and it looks like a mismatch. Look closer and it is one of the more interesting bets in enterprise software right now.

The incumbentFive9 has the scale

Five9 is the boring-in-a-good-way business. Twenty-five years old, listed on the Nasdaq as FIVN, more than three thousand customers, and full-year 2025 revenue of about $1.14 billion, up 10% on the year before. It competes with names like NiCE, Genesys and Amazon for the largest contact centers in the world. When a company with ten thousand agents wants to move to the cloud, Five9 is usually on the shortlist.

What changed the story is AI. In the first quarter of 2026 Five9's core contact-center revenue grew about 8%, which is fine. Its AI revenue grew 68%, to an annual run rate of $125 million. AI went from roughly 8% of subscription revenue a year earlier to about 13%. The company also announced a $200 million buyback, which is the kind of move a business makes when it thinks its own shares are worth more than the market does.

AI agents already handle up to 40% of inquiries. Five9 thinks that could reach 80% by 2029.

In February 2026 the company handed the CEO job to Amit Mathradas, who previously ran Nintex and helped grow the tax-software company Avalara by more than 250%. He arrived with a word to sell: "humantic." The idea is a future where human agents monitor and work alongside AI agents on one screen, instead of AI quietly replacing them in the dark. You can roll your eyes at the coined term. The shape of it is hard to argue with, because it is already happening on the floor.

The challengerUJET has the distribution

UJET was founded in 2015 by Anand Janefalkar, who came out of Motorola and Jawbone. That hardware background matters: instead of treating the smartphone as an afterthought, UJET built its platform around it, letting customers reach support through an app, a website, or a phone number and carry context between them. It spent roughly two years in stealth before launching, which is a long time to stay quiet in a market this crowded.

The move that made UJET matter was not a funding round. It was Google. Google Cloud could have built its own contact-center product from scratch - it builds almost everything else itself. Instead it picked UJET to power its first-party Contact Center AI Platform, wiring UJET into Dialogflow, Agent Assist and the rest of the CCAI stack. In 2023 UJET won Google Cloud's Technology Partner of the Year award for contact-center AI. When your distribution partner is Google, a small customer count stops being the whole story.

UJET raised a total of $76 million in a Series D led by Sapphire Ventures, with GV, Kleiner Perkins, Citi Ventures, DCM and others alongside. In enterprise-software terms that is not a fortune. But UJET is not trying to out-spend Five9. It is trying to be the thing that sits inside a much larger company's product, which is a different and cheaper way to reach scale.

The real fightFrom answering calls to understanding them

The most telling move of the last year was small and easy to miss. In November 2025 UJET acquired Spiral, a conversational-analytics company started by former Amazon engineers. Spiral's job is to take huge volumes of customer conversations and connect them to the specific problems driving them - to tell a company not just how many people called, but why, and what it should fix.

That is the shift worth watching. For most of its history the contact center was about contact management: route the call, resolve the ticket, measure the handle time. The new competition is about customer understanding - turning every conversation into a signal about the product, the pricing, the broken checkout page. UJET's own framing, echoed by industry analysts, is a move from managing contacts to understanding customers. Five9 is pushing at the same wall from the other side, launching voice AI agents meant to replace the old press-one-for-billing phone menus that everyone hates.

Why does the analytics layer matter so much? Because the moat in this business is quietly moving. It used to be owning the phone system. Then it was running everything in the cloud. Now it is the AI agent, and the AI agent is only as good as the conversation data used to train and tune it. Whoever holds the richest, best-labeled record of what customers actually ask for has the advantage. Both companies know it. That is why one bought an analytics startup and the other keeps citing how much of the conversation its AI now handles end to end.

The moat used to be the phone system. Then the cloud. Now it is who understands the conversation.

Two strategiesWhat each is really betting on

Strip away the product sheets and the two companies are making different wagers. Five9 is betting on scale and trust: it already sits inside thousands of large operations, so it can layer AI onto relationships it spent two decades building and grow the AI number without having to win new logos first. Every point of AI adoption inside an existing customer is nearly free revenue.

UJET is betting on being embedded. Rather than winning the enterprise deal head-on, it wants to be the technology other giants ship to their own customers - Google today, potentially others tomorrow. It is the classic challenger move: do not fight the incumbent's distribution, borrow someone bigger's. The risk is obvious, too. When your reach depends on a partner, your fate is partly in their hands.

There is a customer-size angle underneath all of this, too. Five9's strength has long been the biggest, most complex operations, the ones with thousands of agents and compliance teams that ask hard questions. UJET has leaned toward mid-market and digitally native brands that grew up on apps and expect support to feel like the rest of their product. AI blurs that line. As automation handles more of the routine volume, the value shifts to whoever designs the handoff between the bot and the human well, and that is a design problem as much as a scale problem. It is one reason a smaller company can stay in the conversation.

It is worth saying plainly what this is not. Five9 has not bought UJET, and there is no sign it plans to. They are separate companies with separate acquisition strategies - Five9 has folded in businesses like Acqueon, Aceyus and Inference Solutions over the years, while UJET reached for Spiral. They keep showing up in the same comparison charts because buyers keep weighing them against each other, not because a deal is coming.

The takeawayWhy a boring category got interesting

The useful lesson here is not who wins. It is that customer service was the job everyone said AI would automate first, and in the contact center that prediction is quietly coming true. Up to 40% of inquiries handled by software today, a credible path to much more, and a category that most investors ignored now producing some of the fastest AI revenue growth in enterprise software.

If you build or buy software, the pattern is worth stealing. A dull, entrenched category with a repetitive human task and a mountain of conversation data is exactly the kind of place AI reshapes fastest. It can be reshaped by the incumbent with the customers, or by the challenger with the distribution. UJET and Five9 are the live experiment showing both routes at once - and for anyone trying to figure out where AI actually pays, that is a better tell than any demo.

Figures cited from Five9 investor releases, UJET announcements and third-party industry coverage as of August 2026. Comparison graphics are illustrative and use different underlying measures where noted.