MossYC F25 landing page report.
← Back to YC F25
Mosshttps://moss.dev/

Your landing page’s weakest score is Trust.

How Moss ranks

  1. 1.Turbopuffer81
  2. 2.Moss73
  3. 3.Vectorize (Cloudflare)73
  4. 4.Pinecone73
  5. 5.Qdrant73
  6. 6.LanceDB70
Fix Trust
73Growth scoreupdated

category scores

Trust58Mobile web70Information hierarchy76CTA76Value props84
category scoresShowHide
Trust58Mobile web70Information hierarchy76CTA76Value props84

Score over time

9080706050Turbopuffer 80 on Jun 17 (Wayback)Turbopuffer 81 on Jun 24 (Wayback)Turbopuffer 80 on Jul 23 (Wayback)Turbopuffer 80 on Aug 13 (Wayback)Turbopuffer 80 on Aug 23 (Wayback)Turbopuffer 81 on Sep 1 (Wayback)Turbopuffer 81 on Sep 22 (Live)LanceDB 66 on Jun 7 (Wayback)LanceDB 65 on Jun 18 (Wayback)LanceDB 60 on Jul 11 (Wayback)LanceDB 61 on Jul 21 (Wayback)LanceDB 62 on Aug 10 (Wayback)LanceDB 62 on Sep 7 (Wayback)LanceDB 70 on Sep 22 (Live)Vectorize (Cloudflare) 59 on Apr 6 (Wayback)Vectorize (Cloudflare) 73 on Sep 21 (Live)Pinecone 57 on Apr 6 (Wayback)Pinecone 63 on Apr 20 (Wayback)Pinecone 64 on May 5 (Wayback)Pinecone 69 on May 20 (Wayback)Pinecone 72 on May 31 (Wayback)Pinecone 64 on Jun 15 (Wayback)Pinecone 73 on Sep 22 (Live)Qdrant 76 on Jul 3 (Wayback)Qdrant 75 on Jul 14 (Wayback)Qdrant 66 on Jul 24 (Wayback)Qdrant 67 on Aug 8 (Wayback)Qdrant 65 on Aug 24 (Wayback)Qdrant 66 on Sep 12 (Wayback)Qdrant 73 on Sep 22 (Live)Moss 64 on Jul 23 (Wayback)Moss 66 on Jul 31 (Wayback)Moss 73 on Sep 22 (Live)Jul 23Jul 31Sep 22
Past reportsShow past reportsHide past reports
  1. Sep 22, 202673LiveOverall +7 · Mobile web +28 · Information hierarchy +10Viewing
  2. Jul 31, 202666Overall +2 · Information hierarchy +4 · Value props +4View →
  3. Jul 23, 202664First tracked report — strongest Value props 82, weakest Mobile web 42View →

Trust

58Fair

Proof that the product is real, competent, and safe to try.

Moss
Turbopuffer
Pinecone
Trust read
Moss"Backed by Y Combinator" badge above the hero headline
TurbopufferCustomer logo wall (Anthropic, Notion, Atlassian, NYT, Linear) headed "1T+ documents, 10M+ writes/s, and 25k+ queries/s
LanceDB"Tomorrow's AI is being built on LanceDB today" logo strip at the fold's bottom edge
Vectorize (Cloudflare)Headline claim "Everything we learned from powering 20% of the Internet"
PineconeCustomer logo strip: Workday, L'Oréal, Microsoft, Mercado Libre, OpenAI, cohere
Qdrant"30k+ GitHub Stars" badge on the mobile fold

Mobile web

70Fair

The phone experience as a first-class citizen.

Mobile web annotations on the mobile capture1234
  1. Privacy toast overlay — Dead blocker with no CTA overlapping the fold bottom, crowding the primary action zone.
  2. Four-line hero headline — Headline scales down too little, wrapping tall and pushing both CTAs to the fold's edge.
  3. Run a Latency Test — Desktop-proportioned button instead of a full-bleed thumb target at the fold boundary.
  4. Truncated announcement bar — Sticky promo strip cut to 'An AI Agent Liv…' — unreadable message spending first-viewport height.

Information hierarchy

76Fair

Does the design help or inhibit what the page is trying to say?

Information hierarchy annotations on the desktop capture1234
  1. Start Free nav button — Styled at the same black-pill weight as "Run a Latency Test", splitting the fold between two primary actions.
  2. Join us banner — Full-width event promo is the first thing scanned and routes attention away before the value prop is read.
  3. Privacy toast overlay — Persistent notification covers the bottom-right of the fold and competes with the scan path on both captures.
  4. Agent play button — A third dark circular action inside the hero card adds another same-weight click target near the CTAs.

CTA

76Fair

Copy, size, verb, cost-feel — does the primary CTA earn the click?

Moss
Vectorize (Cloudflare)
Qdrant
Pinecone

Problems

  • competes with nav's 'Start Free'
  • 4 words, over the 3-word ideal
  • omits 'free' despite free offer

Value props

84Strong

Title, subtitle, hero — do they sell the product's true value?

Moss’s headline sellsYour Voice AI breaks when retrieval is slow.Get <10 ms search with no vector databases and no latency bottlenecks. Runs directly in browser, edge, device, or cloud.

Working

  • Pain-first headline names a real failure mode
  • <10 ms is a concrete, testable claim
  • "Run a Latency Test" CTA invites proof

Could improve

  • Hero image decorates, doesn't demonstrate
  • No customer or scale proof in the fold
  • Headline omits the <10 ms payoff

Competitors

inferred from product, audience, and value prop
#1Turbopufferturbopuffer.com81
#2Moss73
#3Vectorize (Cloudflare)cloudflare.com73 #4Pineconepinecone.io73 #5Qdrantqdrant.tech73 #6LanceDBlancedb.com70