MMetadata /docs
⌘K
Docs/Tools/query_metadata_analytics_benchmarks

query_metadata_analytics_benchmarks

Query the Metadata.io marketing knowledge base for cross-account industry benchmarks and priors.

Read analytics

Returns data. Calling it changes nothing, so it is safe in an unattended loop.

What it does

Query the Metadata.io marketing knowledge base for cross-account
industry benchmarks and priors.

Arguments

ArgumentTypeNotes
question string required Natural-language question. Should describe a COHORT (industry × size × channel × format / audience), not a specific company.
conversation_id any Optional. Pass the value returned from a prior `_benchmarks` call to keep the query context warm. The two knowledge-base tools never share conversation_ids — pass only IDs returned by `_benchmarks` here.

Request

Request template, not a complete runnable example. Replace placeholders, empty objects, arrays, and zero values with valid account-specific inputs. Read required fields and nested constraints in the full schema. Confirm the account, connected channels, and referenced assets before calling a write tool.

curl
curl -s -X POST https://mcp-server.metadata.io/mcp \
  -H "Authorization: $METADATA_PAT" \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -H "MCP-Protocol-Version: 2025-11-25" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"query_metadata_analytics_benchmarks","arguments":{"question":"<question>"}}}'

Response

Recorded with these arguments. Enum values are the first the tool's own schema declares; ids were fetched live from a list_ call immediately before.

arguments used
{
 "question": "What is the median cost per lead for B2B software?"
}

Real, from the production server, in 21091 ms. Values that identify a customer or disclose money are replaced with typed placeholders; keys, types and nesting are exactly as returned.

recorded response
{
 "status": "ok",
 "tool": "[tool redacted]",
 "sql": "[sql redacted]",
 "text_reply": "[text_reply redacted]",
 "columns": [
  "channel",
  "median_cpl",
  "confidence_tier",
  "experiment_count_distinct"
 ],
 "rows": [
  [
   "facebook",
   "[rows redacted]",
   "HIGH",
   "[rows redacted]"
  ],
  [
   "google_ads",
   "[rows redacted]",
   "HIGH",
   "[rows redacted]"
  ],
  [
   "instagram",
   "[rows redacted]",
   "HIGH",
   "[rows redacted]"
  ],
  [
   "linkedin",
   "[rows redacted]",
   "HIGH",
   "[rows redacted]"
  ]
 ],
 "row_count": 4,
 "conversation_id": "[conversation_id redacted]",
 "message_id": "[message_id redacted]",
 "latency_ms": "[number redacted]",
 "confidence": "HIGH",
 "applied_filters": {
  "account_scope": null,
  "k_anon_floor": true
 },
 "empty_reason": null
}

Other analytics tools:

Browse the tool catalog · machine-readable index