Second, only to Amazon, Facebook Marketplace is among the world’s largest e-commerce platforms, hosting over a billion monthly active users. Deb Liu— former PM for the team that pitched, built, and launched Marketplace — provides an unprecedented look into the product’s development.
The Early Days
Social networking is not just about sharing with friends— in many places, e-commerce is a core part of the “jobs to be done” on Facebook.
- More than 1/2 of people cited Facebook as a place they bought & sold things. Less common in Western markets, this was a blind spot for many at the company.
- The idea was a hard sell, and nothing happened for 5 years, despite hackathons, enlisting interns, and testing new products (eg. Wishlist).
The Launch
The launch of Marketplace wasn’t pretty— a small bug in the fraud queues lead to a massive flood of violating items. Although it was quickly cleaned up, this was just the first in many lessons about what could go wrong on the path to success.
- Focusing on trust.
- Buying and selling meant face-to-face meant trust was important. Users felt more comfortable seeing real identities on the platform— something that isn’t possible on anonymous platforms.
- Integrating with Messenger.
- Integration with Facebook Messenger made communication fast and easy, greatly streamlining the buying and selling process.
- Engaging Facebook’s active users.
- People use Facebook widely and frequently. Adding a local and shipped commerce solution only extended the value of the app.
- Leveraging the Groups community.
- The idea of people-powered commerce— buying from your neighbors —was a part of the ethos of the product from the start.
The Testing
Rotational product manager Bowen Pan joined Facebook and began working on enabling commerce within Facebook groups.
- Many users were members of Facebook Groups communities designed specifically for buying, selling, and trading.
- Early on, commerce in groups was informal and completely organic. Among the biggest issues was identifying groups meant for selling things.
- A feature was added to allow group admins to opt into becoming classified as commerce groups.
- Commerce groups were structured so that items for sale showed up as ranked listings rather than chronological discussion posts.
- The group system was constraining, however. Users had trouble discovering groups, joining them, and finding items. Sellers felt that their reach was limited.
The Pivot
Virtually all two-sided marketplaces struggle with bringing on enough supply and/or demand. Facebook had ample buyers & sellers but failed at connecting them. This left two options:
- Add a tab within the Facebook app.
- Launch a separate app.
- The ability to work within the Facebook ecosystem, where millions were already conducting commerce, made this decision a no-brainer. The feature has remained as a tab to this day because the experience is better integrated into Facebook itself.
- Once the tab had been implemented, “weekly buyer retention” was settled on as the key success metric.
- Buyer engagement among weekly active users remained high throughout the testing period. The product was soon launched in 5 countries, carefully selected to gauge the product’s performance in different areas of the world.
- Some items continued to be sold in groups (eg., marbles, which are mostly bought & sold by enthusiasts). Many sellers, however, wanted to drive maximum discovery to move items.
- Sellers had no incentive to utilize an unknown marketplace. To drive early growth, items that were already being listed in groups could be cross-posted in Marketplace with a single click.