Where are new sellers getting stuck?

Why this matters

Marketplaces are two-sided, but supply usually beats demand to “hard mode.” Buyers come and go on demand-side ads; sellers require a structural commitment (write a profile, photograph inventory, set pricing). If your seller activation funnel leaks, no amount of demand-side spend fixes it — buyers arrive to a thin catalog.

Yet most marketplaces don’t build the funnel. The buyer side gets tons of love (Mixpanel funnels, dashboards, retention cohorts). The seller side gets a seller_signup_completed counter on a CS dashboard somewhere. The result: nobody knows whether the friction is at profile-setup, first-listing, or first-paid-listing.

This recipe puts the per-step funnel in your hand in 5 minutes. Once you know which step bleeds, the supply growth conversation becomes specific instead of “we need more sellers.”

What you get

• A 4-step funnel: seller_signup → profile_completed → first_listing → first_paid_listing
• Per-step drop-off and conversion %
• A published dashboard the supply-growth team can watch weekly

Walk through it

agentry_list_event_names {
  project_id: "default"
}

{
  "analytics_events": [
    { "name": "seller_signup_completed", "count": 8400, "last_seen": "2026-05-15T18:22:00Z" },
    { "name": "seller_profile_saved",    "count": 5240, "last_seen": "2026-05-15T18:21:00Z" },
    { "name": "listing_drafted",         "count": 3120, "last_seen": "2026-05-15T18:20:00Z" },
    { "name": "listing_published",       "count": 1840, "last_seen": "2026-05-15T18:18:00Z" },
    { "name": "listing_first_sale",      "count":  420, "last_seen": "2026-05-15T16:00:00Z" }
  ]
}

agentry_run_recipe {
  recipe_id: "funnel",
  params: {
    project_id: "default",
    steps: ["seller_signup_completed", "seller_profile_saved", "listing_published", "listing_first_sale"],
    window_days: 60,
    conversion_window_hours: 1440
  }
}

{
  "steps": [
    { "name": "seller_signup_completed", "count": 8400, "conversion": 1.000, "avg_time_to_next_s":  3600 },
    { "name": "seller_profile_saved",    "count": 5240, "conversion": 0.624, "avg_time_to_next_s": 86400 },
    { "name": "listing_published",       "count": 1840, "conversion": 0.351, "avg_time_to_next_s": 432000 },
    { "name": "listing_first_sale",      "count":  420, "conversion": 0.228 }
  ],
  "overall_conversion": 0.050
}

agentry_publish_query {
  recipe_id: "funnel",
  description: "Seller activation funnel (60d)",
  params: { steps: ["seller_signup_completed", "seller_profile_saved", "listing_published", "listing_first_sale"], window_days: 60 }
}

{ "publication_id": "pub_seller_act", "public_url": "https://agentry.sh/q/pub_seller_act?key=agp_…" }

The output

Seller activation funnel (60d)

seller_signup_completed       8,400  ──────────────  100.0%
   ↓  1h avg                  3,160  (drop)            62.4%
seller_profile_saved          5,240  ▍                 62.4%
   ↓  1d avg                  3,400  (drop)            35.1%  ★ THE LEAK
listing_published             1,840  ▏                 21.9%
   ↓  5d avg                  1,420  (drop)            22.8%
listing_first_sale              420  ▏                  5.0%

Overall: 5.0% of sign-ups become paid sellers.

Leak diagnosis (profile→listing, 35.1%):
  - Listing creation friction (photos / dimensions / shipping)
  - No nudges after profile save
  - Time-to-publish median 5 days = sellers abandon

Recommendations:
  - Drip emails for "profile_saved but no listing" cohort (D1, D3, D7)
  - Audit listing creation flow (especially photo upload)
  - Per-category listing→sale funnel — diagnose demand-side gap

Public dashboard:
  https://agentry.sh/q/pub_seller_act?key=agp_xxx

Setting it up

You need 4 events fired from the seller-side flow, each with the seller’s distinct_id:

// On the seller-signup completion route
await fetch(`https://api.agentry.sh/v1/analytics/${PROJECT_ID}/`, {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${process.env.AGENTRY_DSN}`,
    "Content-Type": "application/json",
    "User-Agent": "marketplace/1.0",  // REQUIRED — Cloudflare 403s default UAs
  },
  body: JSON.stringify({
    event: "seller_signup_completed",
    distinct_id: seller.email,
    properties: { source: "organic|paid|referral", country: seller.country },
  }),
});

// On the profile-save handler
postAnalytics("seller_profile_saved", { /* ... */ });

// On the listing publish handler (NOT the draft save)
postAnalytics("listing_published", {
  listing_id: listing.id,
  category: listing.category,
  price: listing.price,
});

// On the first-sale webhook from your payments system
postAnalytics("listing_first_sale", {
  listing_id: listing.id,
  order_id: order.id,
  gmv: order.amount,
});

The conversion_window_hours: 1440 (60 days) in the funnel params is important — supply-side activation often takes weeks, unlike a buyer-side checkout funnel that completes in minutes.

Variations

• “Same funnel split by acquisition source — do organic sellers activate at different rates than paid sellers?”
• “Per-country funnel — sometimes activation rate in a launch country tells you whether the playbook is portable.”
• “For sellers who completed profile but never published — pull a sample of 20 emails for a customer call.”
• “Time-to-first-listing histogram. Most sellers either publish in <24h or never — confirm the bimodal.”