Praktika had spent ten months on creator outreach before it found the one that mattered. "I had spent probably like ten months doing this. It was a full time job," says Adam Turaev, Praktika's co-founder and CEO. "I just kept sending those emails, messages to all these influencers and just a few of them just worked out." One of the few was a Brazilian creator named Gustavo, who sent over a video for approval. "I was like, I don't like this video, but, you know, do whatever, like, you can post it."

He posted it. By Adam's count it did "like 10 million views in Instagram, like another 10 million in TikTok, another 10 million in YouTube," and that was the month Praktika went from about $1M ARR to $12M. The video wasn't about the app at all. It showed a coach correcting a learner's mistakes mid-conversation, a format Adam says is now "the main structure of the video" for language-learning apps generally. His instinct had been to ask Gustavo to showcase the product instead. Gustavo's answer, as Adam recalls it, was "believe me."

Paid spend had little to do with it. "We put some budget behind it, but it was like, you know, a tiny budget," Adam says, with first-day costs per trial "literally close to zero." What turned one viral post into a $12M month was replication: "we started just replicating those videos with other creators in other countries as well." Gustavo was hired immediately and still runs Praktika's social media.

The rule behind all of this predates the video. "That was always my, like, rule since the beginning. Like, you know, do not tell influencers to do, like, how to do their job," Adam says. "If you're hiring them because they're like getting an amazing rich, right. So let them do their job." Jacob Eiting's version, from the other side of the table, is that a founder who forces everything through their own approval filter "gonna miss things that are just like, you're blind."

Under 5,000 followers, tens of millions of views

Gustavo came out of a filter, not a shortlist. Adam had no budget for established influencers in Brazil, so he went to a creator discovery platform and set two constraints: a following under 5,000, and views far above what that following should produce. "If you have big following then yeah, you're charging a lot. So I limited this to like less than 5000 followers. But at the same time I filtered out those who are getting like a lot of use these days." The creator it surfaced was doing tens of millions of views on educational videos, had no collaboration email on his profile, and "didn't even realise what he's into." Adam DMed him on TikTok.

Jacob reads it as buying an underpriced asset. Adam pushes back on the word. "Not even about underpriced, but it's more that you're approaching someone on their pick even before they're picking." The point is timing, not price. And he says the filter still works beyond one lucky find: "it can be used as a framework," not targeting micro-influencers as such but "filtering those out who are, uh, hitting like a lot of use with low followings."

How the program runs at $50M ARR

Praktika is now at $50M ARR, "reaching almost like half a million paying subscribers," Adam says, with a team of "a little bit more than 40 people," which puts ARR per employee above $1M. It is profitable. The creator program is past 100 influencers and is still being scaled country by country.

The structure is an influencer lead and country coaches, all of them former creators who were making Praktika videos before they were hired. The lead was making videos in Poland; "she had done an amazing job. And then we hired her as an influencer lead. And she's now running the world programs." Each coach runs "a roster of 20 or 30 influencers, and they just rotate various angles." Knowing what good content is, Adam says, is "the key skill you need to have," because the management side is "not that complicated especially now." The creators stay contractors. Deals are usually a fixed fee plus a bonus on views, but the coaches can restructure for talent: "if there is a good talent just do whatever to to hire this person."

The decision that keeps it working is that paid and organic never share a roster. "This is paid ads and organic content. This creator program that those two are separate streams. We don't mix them up," Adam says. Creators are left to find their own angles because the in-house team eventually runs dry on them, and "every new influencer upcoming is a potential for this." When a video goes viral, it is handed to the paid team, which "reproduce this video with like their own roster of like internal influencers or contractors."

Praktika sold Brazil a job, not a language

Brazil was chosen from a report on English-language penetration: one of the lowest-penetrated large markets, with an income multiplier for English that Adam calls "insane." The warning that came with it was that Brazilians "typically don't pay for consumer subscriptions." Praktika launched worldwide and pointed its distribution at Brazil anyway, because that is where the early adopters were.

Getting a $100-a-year subscription to sell there meant changing what was being sold. "We didn't sell English to them. We sold them an opportunity to get a better job, so it was just a bit of a different angle," Adam says. Every claim in the go-to-market followed, including the "50 x cheaper than a tutor" line he says "everybody uses now." David Barnard's generalization for anyone picking a problem: "look for a problem where you're selling money."

The avatars that worked in ads and failed in the app

Adam's biggest fail of the past year was moving a growth idea into the product. Unhinged, cursing avatars "were going viral in Upper Funnel," so the team added them to the app. It made more people unhappy than happy, and they didn't understand it. "The image is fine. Like, you know, I want to talk to this avatar, but why is it cursing, right?" The principle it clarified: "we're not serving AI companions. We're serving AI coaches."

Praktika now draws the line explicitly. Product has clear, agreed boundaries on "a one sheet of description." Growth has almost none. "We let influencers to create and to come up with like pretty much any angle that they want," Adam says, and the difference between the two sides "is pretty dramatic."

In the full episode, Adam also explains why Praktika's biggest win of the year was moving from Flutter to native iOS, why he builds for the best model quality and trusts inference costs to fall, and why fear, not skill, is the reason the app uses avatars at all.

Guest links