We're sharing a guest blog post from Sonia Nagar on key themes and learnings for marketplaces in 2025. This was previously shared on Sonia's Substack here and also as a post in the community here.
‍
In 2025, one marketplace story stood out above all: Mercor’s meteoric rise. What started in 2023 as an AI-driven hiring platform quickly became one of the fastest-scaling talent marketplaces, hitting an estimated $500 million in annualized revenue and securing a $10 billion valuation after a $350 million Series C round. That growth underscores a broader shift in how marketplaces create value in the AI era. Winning platforms shrink time-to-value from weeks to minutes, use AI to create synthetic supply when real supply lagged, and slash switching friction to near zero. Reputation systems have been broken wide open too, trust became portable and real-time, not earned only over years. Below is more on five marketplace insights from 2025, illustrated with real-world examples of these trends in action.
‍‍

1. Time-to-Value Is the New Kingmaker

AI has collapsed onboarding friction, and the companies winning today give users value nearly instantly.
‍
Example: Instawork’s "Instant Shifts" for Hourly Labor‍
‍
In hourly labor marketplaces, the traditional timeline looked like:
sign up → profile approval → job search → application → interview → first shift → earnings.

That could take days—or weeks.

Instawork uses AI-based credentialing, profile automation, and skill verification to compress that journey into hours. New workers can now:

• Auto-generate a complete profile from a résumé or driver’s license
• Get instantly matched to shifts based on skills, location, and reliability signals
• Start earning the same day

‍

2. Negative Network Effects Show Up Sooner

With AI agents generating content and users switching tools faster than ever, noise and low-quality supply surface earlier.

Example: Amazon Marketplace’s AI-Driven Product Flood

Amazon’s flywheel has always been "more sellers + more SKUs = better selection = more value."

But in 2025, AI made it effortless for sellers to:

• Spin up auto-generated product designs
• Create AI-written listings at scale
• Pump out synthetic reviews and Q&A

Instead of increasing choice, this caused listing overload, lower average product quality, higher return rates.

Shoppers couldn’t distinguish real brands from mass-produced AI junk, and conversion rates on broad search terms actually declined—a classic negative network effect:

Mitigating strategies include:

• Stricter brand and identity verification
• AI-content filtering for listings and reviews
• Greater emphasis on "Amazon’s Choice" and vetted storefronts
• More prominence for trusted brands and repeat sellers

In the AI era, marketplaces can’t wait for scale to fix quality. They have to defend trust on day one, or growth turns toxic.
‍

3. Synthetic Supply Is Now a Real Bootstrapping Strategy

AI-generated supply is transforming how networks reach liquidity.

Airbnb has used AI-generated listing descriptions and photos to improve conversion and "seed" the impression of high-quality inventory.

• Etsy launched tools that auto-generate product photos and descriptions, increasing the supply surface area without relying solely on seller effort.
• In B2B, G2 and Capterra use AI to generate structured summaries of reviews, making category pages feel "full" even with fewer contributors.

This shifts the playbook from "grow supply first" to "generate supply, attract demand, then convert to real supply."
‍

4. Reputation Layers Are Becoming Portable

Trust is escaping the platforms where it was born.

• Shopify’s Shop app now carries a merchant and delivery reputation across their entire ecosystem—not just one store.
• Stripe Identity is quietly becoming a cross-platform trust layer for marketplaces and fintech apps.
  ‍

5. Winner-Take-Most Is Weakening

AI-driven multi-homing is eroding single-platform dominance.

• Google’s SGE and AI shopping agents let users compare Amazon, Shopify stores, and brand sites in one query.
• Uber, DoorDash, and Instacart now compete inside AI-driven comparison layers, where the "agent" picks the best option for the user.
• Travel agents (Hopper, Expedia, Booking) increasingly lose exclusive control as AI tools auto-shop across all of them simultaneously.

Owning workflows, proprietary data loops, and trust infrastructure matters more now than ever for marketplaces to keep buyers and sellers locked in.‍

Let me know if you have any takeaways or even learnings in the comments below.
‍
You can connect with Sonia to discuss this post in the Everything Marketplaces community here. A big thanks to Sonia for also being active in the community, where she is often sharing her marketplace experience, insights, and helping earlier stage founders.