Three months ago, my Spanish was basically a ghost from school: years of classes, almost nothing usable, and no real ability to hold a conversation. So I gave myself 90 days to change that and set one clear ending for the story: a real 30-minute conversation with a Colombian teacher.
Watch on YouTube: Main challenge video (link placeholder)
Watch on YouTube: Full 30-minute conversation (link placeholder)
Spanish had been sitting in the back of my mind for years. I had studied it in school, but I was one of the worst students in the class, and by the end of high school I still could not do the one thing that actually matters: talk to someone. After that, I barely touched the language for more than a decade, so whatever weak foundation I had was gone.
That is why this challenge mattered to me. It was not really about picking a random language. It was about going back to something that had ended badly and seeing whether a focused, sensible plan could turn years of failure into an actual conversation.
There is one important caveat in this story: I am a native French speaker, so Spanish is not equally difficult for me as it would be for everyone. For many English speakers, reaching the same point may take longer. Even so, Spanish is still one of the more approachable languages for English speakers, which made it the right language for a short, high-pressure experiment.
According to the Foreign Service Institute (FSI), Spanish is a Category I language — the easiest category for native English speakers, estimated at around 575–600 class hours to reach professional working proficiency. That is roughly 23–24 weeks of full-time study, placing it among the shortest timelines of any language. For reference, here is the full FSI classification table:
The FSI estimate targets full professional proficiency — being able to negotiate, give a presentation, or read a contract in Spanish. This challenge aimed at something much narrower: a real 30-minute conversation. To get there in three months on 30–45 minutes a day, I leaned hard on the 80/20 rule — focus on the 20% of the language that drives 80% of everyday conversation: the most frequent 2,000 words, the handful of tenses people actually use when speaking, and the patterns that come up again and again. The FSI's 600 hours is honest about the full destination; the challenge is honest about a much smaller, more specific one.
I wanted the challenge to look like real life, not like a language boot camp. So I capped active study at about 30 to 45 minutes a day. If I watched something in Spanish, I treated that as extra input, but even then I tried to keep it reasonable, usually around an hour. Nothing about this setup was meant to resemble an immersion retreat. The whole point was to find out what could happen inside the kind of schedule a working person might actually sustain.
The ending was fixed from the start. At the end of the three months, I would get on a call with a Colombian teacher and see what was really there.
I started with something that sounds strange until you try it: I wrote short speeches and useful sentences, then memorized them. I focused on topics I knew would probably come up later: who I am, where I am from, what I do, my YouTube channel, and how I was learning Spanish.
I needed a textbook, but I did not want one that would bury me under too much grammar too early. Assimil was light enough to fit the challenge and structured enough to keep me moving forward one lesson at a time.
Note: I am not affiliated with Assimil at all. I just like their courses.
Then came vocabulary. I built an Anki deck around high-frequency Spanish words, with translations, example sentences, and audio. The original target was 2,000 cards, but the clock forced me to stop at around 1,000.
For speaking practice, I mostly used Island Koala. I should be transparent here: this is my own tool, so I am obviously biased. What I found genuinely useful was the immediate grammar feedback, the option to practice by voice or text at any time, and the ability to capture vocabulary and push it into Anki. But this part of the method is not tied to my app. You can absolutely use another AI tutor and still run the same system.
And finally, I added native input through Narcos. At the beginning, it was far above my level. I did not understand much. But it was interesting enough that I kept coming back, and over time it became less like noise and more like a real language.
None of these tools did the job alone. The speeches gave me something to say, the textbook gave me structure, the deck gave me words, the tutor gave me feedback, and the show gave me contact with real speech. The story of the challenge was really the story of those five things reinforcing each other. Island Koala happened to be my tutor in this run, but any AI tutor that gives fast corrections and consistent speaking practice can fill that role.
I used iTalki to find native tutors, and around the halfway point I booked a real one-hour lesson to see whether my progress was real or whether I was just getting comfortable inside my own system. The answer was both encouraging and humbling. I could communicate more than before, but my vocabulary was not carrying me far enough. Too many unknown words were still slowing everything down.
Up to that point, I had been relying too much on general exposure, structure, and conversation practice.
After that lesson, the strategy became clearer: I needed more deliberate vocabulary work, not just more time around the language.
So I started capturing unknown words from Island Koala sessions and pushing them back into Anki for focused review. If you use another tutor, the same loop still works: capture unknown words and review them deliberately.
In the last week, the challenge started to feel very real. I went back over the speeches and sentence patterns I had memorized and reviewed the topics that were most likely to come up: introducing myself, talking about Tahiti, describing my work, explaining the channel, and telling the story of the challenge itself.
At that stage, I was not trying to cover everything. I was trying to walk into the conversation with enough language ready at hand that I would not freeze when the obvious subjects appeared.
When the final call happened, I was happy with the result. Not because the Spanish was perfect, but because it was real. I could hold a conversation, move through general topics, and stay in the language. My own estimate at the end was around B1.
The weaknesses were also obvious. Fast speech was still difficult. Background noise would make things worse. Group conversations would be harder. And vocabulary was still the main bottleneck, especially once the conversation moved away from familiar themes.
If there is one lesson that runs through the whole story, it is this: people underestimate active vocabulary study. Input matters. Speaking matters. But if you do not know enough words, everything slows down. That became clearer to me with every week of the challenge.
Some links on this page may be affiliate links. If you purchase through them, I may earn a small commission at no extra cost to you. I only link to tools I have personally used and found genuinely useful.
This is the same Spanish deck I used during the challenge, including native proofreading and multiple pronunciation options.