Modern examples of how companies got their first few customers

First customers

Edwin Chen's approach was very straightforward: built the v1 in a couple of weeks, posted on his data science blog, and became profitable in month one. Early customers include Twitter, Twitch, and Airbnb (in the search, recommendations, and content moderation space).

At Twitter, he had this problem when trying to build a sentiment classifier from 10k labeled tweets - positive or negative. But their human data labeling system was just two people they hired off of Craigslist working 9 to 5. Starting required waiting a month.

At Facebook, he tried to create a dataset of 50k labeled businesses to distinguish between different business types. They hired an outside firm, which took 6 months to deliver it, which on top of that was riddled with errors - restaurants mislabeled as coffeeshops, coffeeshops as hospitals.

Factors

• He already had a data savvy audience from previous technical blog posts. When he posted about Surge, exactly the right readers were there.
• Edwin was an expert on this problem. He had spent most of his career solving data problems at big tech companies.
• Timing: Edwin launched Surge around the time GPT-3 launched.

First customers

Yasser Elsaid's got his first customers on X. When GPT-3 API just launched, he hustled by quickly creating free custom chatbots for books and podcasts in relevant communities (Reddit, Twitter), hiding the pricing page for Reddit users to avoid spam flags and maximize credibility.

He had a breakthrough in February 4, 2023 when his demo video exploded overnight (from an account with only 16 followers), gaining 1,500+ likes and thousands of users. At launch, the product was completely free. Business owners demanded a pricing page, and within two hours of launching it, Elsaid had his first paying customer at $10/month.

From there, he quickly hit revenue milestones:

• $10K MRR by March 2023
• $64K MRR by May 2023
• $1M ARR in early 2024
• $5M+ ARR by February 2025

Factors

• Perfect timing: Chatbase launched during peak ChatGPT hype, capturing businesses hungry for AI automation.
• Built one of the first "Chat with PDFs", which contributed to virality
• Building in public on X allowed AI influencers to amplify growth organically.

First customers

Despite having no previous experience in legal, Max got his foot in the door by reaching out to to hundreds of law firm partners on LinkedIn DM. He offered to pay their hourly rate - most waived it and still showed up, appreciating the respect for their time.

His earliest win was when he convinced a top Swedish law firm to let him co-work at their offices. In return he would build prototypes for them, with the ultimate goal of having their firm adopt an AI product. He would badger their lawyers with questions at lunch time, demoing the product and testing its usefulness. This tight feedback loop (getting direct feedback, in person, every day) was key to building a product that legal firms wanted.

In a short amount of time, Legora expanded its user base to more than 250 legal firms across 20 markets by May 2025, including major clients like Bird & Bird, Cleary Gottlieb, and Goodwin.

Factors

• Got his foot in the door by DMing partners on LinkedIn and respecting their time.
• Embedded himself at a law firm to create a tight feedback loop between product and customer.
• Gave value back to early customers by sharing product ideas and AI possibilities to get people excited.