How to Launch (Again and Again)

Mañalac argues that launching is not a single high stakes event but a repeated process as a startup ships products and features. A team that spends six months polishing its first launch may run out of time before it gets another chance. Launching early also helps founders practice their pitch and learn how people respond.

A simple landing page can introduce the company, explain the product, provide contact details, and invite people to act. Mañalac recommends testing an idea with friends and family, then moving quickly to strangers who better match the intended users. Lugg tested demand before its app worked by helping IKEA shoppers with a truck, confirming that moving their purchases was an urgent problem.

YC founders can test products with peers on Bookface or the Startup School forum, which Mañalac says had about 40,000 founders. Magic shared a basic text based service that spread to Reddit and Hacker News, bringing 40,000 signups over a weekend, though she describes this as an unusual outcome. For community posts, she advises founders to learn each site's rules, write naturally, explain what they learned, and invite relevant questions.

A waitlist can help manage demand and encourage referrals, as Magic and Superhuman demonstrated. Joy contacted more than 50 wedding bloggers and got four responses that meaningfully contributed to early growth. Stripe repeatedly launched products through its blog, Hacker News, social media, and press, while Glossier used a regular product cadence and repeated its launch activities every six to eight weeks.

Symetra founder Gaddy Avron kept an email list of people he met while discussing the company, then asked them to share its TechCrunch story. Mañalac recommends that founders start collecting contacts and sending occasional updates before they have a product. Her closing advice is to keep testing launches throughout a company's life rather than reserving effort for one big day.