CloudTalk Bets the Phone Call Isn't Dead - It's Just Dumb

For most of the last decade, the smart money in customer software said the phone call was finished. Chat was cheaper. Email scaled. Bots never slept. Then a company you have probably never heard of, run out of Bratislava, quietly started handling hundreds of millions of business calls a year - and raised $28 million on the argument that everyone had misread the problem.

The problem was never the call. It was the software behind it. That is the whole CloudTalk thesis in one sentence, and it explains why a Slovak startup keeps turning up on 2026 comparison pages next to UJET, Mitel and Salesforce - companies with far bigger names and far bigger addresses.

CloudTalk was founded in 2016 by Martin Malych and Viktor Vanek. It did not launch as a grand vision for the future of voice. It launched as a click-to-call widget - a single button you could drop on a website so a visitor could reach a real person. Small idea. But the founders kept pulling the thread: if you own the button, you own the call, and if you own the call, you own everything that happens on it.

01 / THE PITCHThe unsexy market nobody wanted

Contact center software is not a glamorous category. It is routing rules, hold music, area codes, and agents reading from scripts. Which is exactly why it was such a good place to build. Boring markets hide good businesses, because the incumbents get comfortable and the talented founders go chase something shinier.

CloudTalk's move was to treat the phone call like software instead of infrastructure. That reframe sounds academic until you see what falls out of it. If a call is software, you can spin up a local number in 140+ countries without shipping any hardware. You can route it, record it, tag it, score it. And once every call is data, you can point AI at it.

That last number is the one CloudTalk likes to translate into human terms: more than 400 million conversations is roughly 665 years of continuous talking. It is a marketing line, but it is also a real moat. Every one of those calls is training material and evidence that the routing, the uptime and the audio quality actually hold up at scale.

02 / THE PRODUCTWhat AI actually does to a phone call

Here is where CloudTalk gets interesting, and where the demo-driven hype around AI mostly falls apart. The company did not build a robot to replace the agent. It built AI to erase the paperwork around the agent. When a call ends, the software writes the summary, files the notes, tags the outcome and reads the mood of the conversation. The agent moves to the next call instead of spending five minutes typing.

Multiply that by a support team's daily volume and the pitch stops being abstract. An agent who saves an hour of admin a day is an agent who takes more calls, or takes them less exhausted. The product is not really the phone. It is the memory of the call - who said what, how they felt, what happens next - turned into structured data that flows into the CRM automatically.

None of this is science fiction, and that is the point. The unglamorous work - summaries, notes, sentiment, scoring - is where AI actually lands in customer service in 2026. CloudTalk bet its roadmap on the boring middle of the funnel, and the boring middle is where the hours hide.

03 / THE FIGHTCloudTalk vs UJET vs Mitel vs Salesforce

Look up any call center comparison this year and you get the same four-way stare. Each player owns a different piece of the conversation, and CloudTalk has quietly wedged itself into the frame.

UJET pitches itself as the platform built for the smartphone era, leaning on app-native experiences. Mitel comes from the deep enterprise telephony world - the wiring, the on-prem heritage, the big deployments. Salesforce, now branding its service product as Agentforce Service, wants the call to live inside the CRM it already owns, wrapped in its own AI. CloudTalk's angle is speed and reach: fast setup, international numbers, strong CRM integrations, and voice AI aimed squarely at small and mid-market teams that do not have a telecom department.

That positioning is why the company can punch above its funding. It is not trying to out-enterprise Mitel or out-CRM Salesforce. It is trying to be the fastest way for a growing team to get a smart, well-documented phone system running by Friday. In a market where the incumbents sell six-month implementations, "by Friday" is a strategy.

04 / THE PLAYBOOKBuilt in Europe, sold everywhere

There is a second story tucked inside the first one, and it is worth stealing. CloudTalk did all of this without a Silicon Valley zip code. Its offices sit in Bratislava, Prague, Toronto and Barcelona, with people in London, New York, Mexico City and Munich. Around 140 employees across more than 20 nationalities. The $28 million Series B came from European funds - co-led by KPN Ventures and Lead Ventures, with Point Nine Capital, henQ, Presto Ventures and Orbit Capital in the round.

The lesson for anyone building outside the usual hubs: geography is not the moat people think it is. A calling product is global by default, distribution runs through integrations and search, and customers do not ask where the servers' owners eat lunch. CloudTalk grew one boring feature at a time until the comparison sites had no choice but to list it.

• 2016Founded in Bratislava by Martin Malych and Viktor Vanek as a click-to-call widget.
• 2018Grows into a full cloud call center with local numbers across 140+ countries.
• 2021Raises a Series A (reported around €6M) from European venture backers.
• 2023Ships AI features - call summaries, automated notes, sentiment analysis.
• 2024Closes a $28M Series B co-led by KPN Ventures and Lead Ventures.
• 2026Sits on the year's CCaaS shortlists beside UJET, Mitel and Salesforce.

05 / THE BETWhy the human voice still wins

The contrarian core of CloudTalk is a fact the AI wave made everyone forget: when a customer problem is genuinely hard, expensive or emotional, people still want to talk to a person. The chatbot handles the password reset. The phone call handles the cancellation, the complaint, the six-figure renewal. CloudTalk's bet is that the best AI in customer service does not remove the human from the line - it makes the human on the line better prepared, better documented, and less buried in admin.

You do not have to believe voice will beat every channel to see the opening. You only have to believe that the phone call was never going to disappear, and that whoever made it smart first would end up owning a valuable, unfashionable corner of the software world. That is the corner CloudTalk has been quietly furnishing since 2016.

Whether it out-lasts UJET, out-scales Mitel or out-maneuvers Salesforce is an open question. But the company has already proven the thesis that got it funded: the phone call was never dead. It was just running on old software, waiting for someone to rebuild it.