Finding Demand for a New Event Format
The useful output wasn't a topic cluster. It was an event the business could run next week.
Result: foot traffic doubled within one quarter of launching the language-exchange nights, and the format stuck.
The problem
A cafe needed a reason for people to come back on a schedule. The question was which event had real local demand but few places offering it, and whether the business could try it without committing to a big program first.
What I did
I scraped Reddit and Meetup for how local people talked about social events, language learning, and recurring meetups. I also joined public WhatsApp groups, downloaded the chat history, and ran n-gram, term-frequency, topic, and sentiment analysis to see which events people mentioned and reacted to. I compared three concepts: language exchange, karaoke, and trivia. The tools were NLTK, spaCy, and Hugging Face models, and I trained NLP models on the combined corpus.
Language exchange stood out as underserved: steady interest and few events offering it. So I recommended a small test: run language-exchange nights, watch attendance, and compare the next quarter with the one before.
How the business used it
The finding changed what the business sold. Marketing moved from "come visit the cafe" to a scheduled social reason to show up. The copy got casual, energetic, and multilingual, along the lines of "Welcome to the worst language class you'll ever take and the best party you'll find this week."
The events also gave the business a repeatable unit for testing. I used designated holdout and treatment days to compare event formats, audiences, and marketing approaches.
The work fed into a pivot and an asset sale. The new owner turned the business into an event-based lounge and kept me on for attribution, pricing, and performance tracking. My later analysis for the lounge found karaoke and trivia under-offered with strong interest, and it now runs language exchange and trivia nights.