Multi-touch attribution tracker. Rebuilds the full visitor journey before a form submission — without a CRM, without a dashboard.
The interface is your AI agent.
Trail drops a lightweight script (~3kb) on your site. It tracks every session a visitor has before converting, stores the channel for each one, and links them to a lead ID when a form is submitted.
Session 1 — google_ads (clicked a Google Ad)
Session 2 — direct (came back directly)
Session 3 — google_profile (clicked the Google Business link)
↓
form submitted → trail_vid injected → lead linked to all 3 touchpoints
Query the data through an MCP server — ask your AI agent "where are my leads coming from?" and get a real answer.
| Signal | Channel |
|---|---|
gclid |
google_ads |
fbclid |
facebook_ads |
li_fat_id |
linkedin_ads |
ttclid |
tiktok_ads |
utm_source=X |
exact value (e.g. google_profile, newsletter) |
| Referrer = search engine | organic_search |
| Referrer = social network | organic_social |
| Nothing | direct |
Click IDs (gclid, fbclid…) take priority over utm_source. UTM parameters and click IDs are persisted in localStorage with a 30-day TTL — if a visitor lands via a Google Ad, closes their browser, and returns directly a week later, the original channel is still attributed at conversion time.
Requires Node.js 22+.
DATABASE_URL=postgres://... npx -y --package=@silverbackbase/trail trailWithout DATABASE_URL, Trail starts on http://localhost:3000 with a SQLite database created automatically at ./trail.db.
Add the tracker to your site:
<script src="http://localhost:3000/t.js"
data-account-id="your-site"
async defer></script>
git clone https://github.com/silverbackb/Trail.git
cd trail
pnpm install
pnpm --filter @trail/tracker run build
pnpm --filter @silverbackbase/trail run buildRequired environment variables:
| Variable | Description |
|---|---|
TRAIL_URL |
Public URL of your Trail instance (e.g. https://trail.yoursite.com) |
DATABASE_URL |
PostgreSQL connection string. If set, Trail uses Postgres. If absent, falls back to SQLite. |
DB_PATH |
SQLite file path (default: ./trail.db). Only used when DATABASE_URL is not set. |
TRAIL_TZ |
Timezone for MCP output (default: Europe/Paris) |
Storage: Trail automatically picks the right adapter at startup — PostgreSQL when DATABASE_URL is set, SQLite otherwise. No migration needed: if PostgreSQL is empty on first start and a SQLite file exists, Trail migrates the data automatically.
If you use Google Tag Manager, create a Custom HTML tag with two script blocks:
<script>
window.trailConfig = {
accountId: "your-site"
};
</script>
<script src="https://trail.silverbackbase.com/t.js" async defer></script>Set trigger to All Pages.
Trail automatically detects form submissions. At submit time, it injects a hidden trail_vid field into the form — your backend, CRM, or webhook receives it alongside the other form fields, no extra code required.
To manually trigger a conversion (e.g. from a custom flow):
fetch('https://trail.silverbackbase.com/convert', {
method: 'POST',
headers: { 'Content-Type': 'application/json' },
body: JSON.stringify({
visitor_id: document.cookie.match(/trail_vid=([^;]+)/)?.[1],
account_id: 'your-site',
lead_id: 'contact@example.com',
}),
});Trail exposes an MCP server so your AI agent can query attribution data directly.
Add to your .mcp.json or claude_desktop_config.json:
{
"mcpServers": {
"trail": {
"command": "npx",
"args": ["-y", "--package=@silverbackbase/trail", "trail-mcp"]
}
}
}The MCP server connects to the same SQLite database as your local Trail server.
Point your MCP client to your Trail instance:
{
"mcpServers": {
"trail": {
"type": "http",
"url": "https://your-trail-instance.com/mcp",
"headers": {
"Authorization": "Bearer <your-token>"
}
}
}
}Using the hosted version? → silverbackbase.com
Available tools:
| Tool | Description |
|---|---|
trail_create_account |
Create a client account, get the tracker snippet (header or GTM) |
trail_list_accounts |
List all accounts |
trail_get_recent_sessions |
Check that the tracker is receiving visits (no conversion required) |
trail_list_leads |
List leads with acquisition channel and conversion date |
trail_get_journey |
Full touchpoint journey for a lead |
trail_get_report |
Attribution report by channel (first / last / linear) |
trail_get_top_paths |
Most common multi-touch paths before conversion |
trail_get_channel_performance |
Visitors / leads / conversion rate per channel |
By default the tracker points to trail.silverbackbase.com. Brave and uBlock Origin block requests to third-party domains. A reverse proxy routes tracking through your own domain — invisible to blockers.
Next.js (next.config.js):
module.exports = {
async rewrites() {
return [
{ source: "/api/t.js", destination: "https://trail.silverbackbase.com/t.js" },
{ source: "/api/t", destination: "https://trail.silverbackbase.com/t" },
];
},
};Then in the snippet, add data-api-url="" so the tracker uses the current domain:
<script src="/api/t.js" data-account-id="..." data-api-url="" async defer></script>Cloudflare Worker:
export default {
async fetch(request) {
const url = new URL(request.url);
url.hostname = "trail.silverbackbase.com";
return fetch(new Request(url, request));
},
};Nginx:
location /api/t {
proxy_pass https://trail.silverbackbase.com;
proxy_set_header Host trail.silverbackbase.com;
}packages/
tracker/ Browser script — visitor ID, session detection, channel capture (~3kb)
server/ Node.js server — HTTP API + MCP + SQLite / PostgreSQL
- Tracker — vanilla TypeScript, compiled to ~3.5kb IIFE via esbuild
- Channel persisted in
localStorage(30-day TTL) — cross-session attribution preserved - Scroll depth (
scroll_depth_pct) and time on page (time_on_page_sec) captured at conversion
- Channel persisted in
- Server — Hono +
@hono/node-server - Database — SQLite via
node:sqlite(built-in Node 22+) for self-hosted; PostgreSQL viapostgresfor cloud deployments - MCP —
@modelcontextprotocol/sdkwith Streamable HTTP transport
MIT — part of the SilverBackBase ecosystem.