Distribution Intelligence

Pilot

Your network's retention risk, answered inside your own systems.

Four allowlisted datasets from your own distribution workspace: the roster you manage, current churn risk, the retention actions logged against it, and an adviser's own record. Read only, inside a signed data contract. A gated pilot in Czechia today.

Gated pilot. Access is granted per workspace, so this one starts with a conversation.

Talk to us about delivering this Connect it to your AI

Where it runs today: your own AI client calls it as a tool over MCP.

What changes

Who inside your organisation is better off, and what they can show for it.

Head of Distribution at an adviser network
Asks for the team's risk mix in one sentence, and gets counts by critical, high, medium and low, plus how many are flagged act now.
Capability manifest and distribution pilot connector: the team_churn dataset returns a roster summary with total, critical, high, medium, low, average churn, average tenure and act_now.
Regional manager who owns a team of advisers
Starts the week with the roster ranked worst first, each line carrying risk level, churn score, tenure and an act now, review or monitor flag.
The team_churn dataset ranks every roster adviser highest risk first and stamps the snapshot date on the answer. Pilot, gated, Czechia only, 5 September 2026.
Retention lead at a distributor
Reads back the calls, emails, meetings and tasks already logged against the roster, with status and due date, so a handover does not lose the history.
The retention_actions dataset returns action type, title, status and due date, newest first. Write tools are deliberately not reachable, so the AI route reports and never logs.
Data owner or COO who has to sign the connection off
Sees exactly which four datasets are exposed, that access is read only, and that anything outside the contract allowlist is refused.
Manifest permissions verified_workspace and distribution:read, access read, data class aggregate, allowlist of exactly team_roster, team_churn, retention_actions and advisor_profile. No dataset in the allowlist has an email or phone field.
Head of Operations building an internal assistant
Puts the retention answer inside the assistant managers already use, instead of shipping one more dashboard for them to log into.
The only declared execution surface in the manifest is MCP, so the capability is reached as a tool by an AI client, not as a screen.

Two ways to get it

Have it delivered, or let your own AI call it.

As a service

We deliver this with you, not through a button. We agree the dataset allowlist, put it in a signed data contract, verify the workspace, load the roster you actually manage, and hand back the risk mix, the ranked roster and the retention log as a reading your managers can act on. Access stays approval gated while per workspace authentication is finished, so every pilot starts with a conversation. Write to us with the country, roughly how many advisers you manage, which datasets you would allow, and who on your side signs the data contract. A person replies.

Tell us what you need

One message, no form. We answer with what it would take.

Through your own AI

One tool, query_distribution, takes your question and a dataset name and reads one of four allowlisted datasets: team_roster, team_churn, retention_actions or advisor_profile. Your assistant can rank the roster by risk, read back the retention log, and answer the questions listed above with the snapshot date attached. It is read only, so it can report but never change a roster or log an action. Nothing is open today. The connection is approval gated, the pilot proxy stays off until a workspace contract is signed, and per workspace authentication is still being wired, so no customer AI can call it yet. Ask for pilot access and we will tell you when your workspace can connect.

Connector: https://api.emable.ai/mcp
Tools it adds: query_distribution

Ask for access How the connector works

What your AI can answer

Connect it once, and your assistant answers these.

Questions in your own words, each one landing on a published tool. Ask us and we answer the same ones for you.

1How many advisers are we tracking in this workspace right now? | tool: query_distribution, dataset team_roster?query_distribution
2Which firms do the advisers on our roster currently sit with? | tool: query_distribution, dataset team_roster?query_distribution
3How does retention risk look across the whole team today? | tool: query_distribution, dataset team_churn?query_distribution
4Who is at the highest risk of leaving, ranked worst first? | tool: query_distribution, dataset team_churn?query_distribution
5How many of our advisers are flagged act now rather than review or monitor? | tool: query_distribution, dataset team_churn?query_distribution
6What is the average tenure and the average churn score across the team? | tool: query_distribution, dataset team_churn?query_distribution
7What retention actions have we logged, and which of them are still open? | tool: query_distribution, dataset retention_actions?query_distribution
8Which logged retention tasks are past their due date? | tool: query_distribution, dataset retention_actions?query_distribution
9Which act now advisers have no retention action logged against them yet? | tool: query_distribution, datasets team_churn and retention_actions?query_distribution
10How fresh is the churn snapshot these scores come from? | tool: query_distribution, dataset team_churn?query_distribution
11For the adviser record claimed by this account, which product licences are active and what does the churn snapshot say? | tool: query_distribution, dataset advisor_profile?query_distribution

Distribution Intelligence answers these through the tools above. A question outside them is refused, not guessed.

What it rests on

The evidence, and the edges.

Evidence

Capability manifest: status pilot, exposure gated, Czechia, provider emable, access read, permissions verified_workspace and distribution:read.
Dataset allowlist, four entries: team_roster, team_churn, retention_actions, advisor_profile. Source: the emable capability catalogue.
Churn fields returned per adviser: risk level, churn score, tenure in months and an act now, review or monitor flag, with a roster summary and the snapshot date on every answer.
Charging: one query within the signed workspace contract. Contract priced, with no published credit ceiling.
Evidence freshness: defined by the signed workspace data contract. Manifest, 5 September 2026.
Verified usage: Not enough verified runs yet. Public statistics appear only after 20 verified runs across 5 different accounts.
No reliability grade is published for this capability, and no verified customer figures exist for it.

Defined by the signed workspace data contract.

What it does not do

Access is restricted to verified workspaces.
The pilot remains approval-gated until shared OAuth is complete.
Read only. Adding an adviser to a roster or logging a retention action is not reachable from the AI route.
Only four datasets are readable: team_roster, team_churn, retention_actions and advisor_profile. Anything outside that allowlist is refused.
The pilot connector is off by default. Where it is switched on it runs against a shared demo workspace, and every answer is labelled pilot data that is not customer specific.
A churn score is a model signal over register and tenure data. It is not a statement that a named person is leaving.
No adviser income data exists here, and no dataset in the allowlist has an email or phone field.
Not enough verified runs yet, so no usage statistics are shown on this page.

Verified usage

Not enough verified runs yet

Public statistics appear only after 20 verified runs across 5 different accounts.

The contract, for the people who will wire it

Status, permissions, tools, schemas and where it runs. The same values the API returns, so nothing here can drift from what your client will actually get.

Status, access and price

Status: Pilot
gated
Provider: emable
Manifest version: v1.0.0
Countries: CZ
Data class: aggregate
Access: read
Permissions: verified_workspace
distribution:read
Charging unit: One query within the signed workspace contractPricing model: contract. Agreed in the engagement.Credits are the platform's metering unit, not a price. We quote the engagement.

Tools and schemas

Query distribution data

read

query_distribution

Queries an allowlisted aggregate distribution dataset.

Returns

A workspace-scoped aggregate result.

Detailed guidance for this tool is not published yet.

Input schema

{
  "type": "object",
  "properties": {
    "question": {
      "type": "string",
      "minLength": 3,
      "maxLength": 1000
    },
    "dataset": {
      "type": "string",
      "minLength": 1,
      "maxLength": 80
    }
  },
  "required": [
    "question",
    "dataset"
  ],
  "additionalProperties": false
}

Output schema

{
  "type": "object",
  "properties": {
    "summary": {
      "type": "string"
    },
    "aggregates": {
      "type": "object"
    },
    "result_reference": {
      "type": "string"
    }
  },
  "required": [
    "summary",
    "aggregates"
  ],
  "additionalProperties": false
}

Surfaces and connectors

mcp

No standalone MCP URL is deployed for this capability yet.

Specialist workspace, sign in required

Sources, limits and support

Sources

Datasets allowlisted in the signed provider workspace contract.

Limitations

Access is restricted to verified workspaces.

The pilot remains approval-gated until shared OAuth is complete.

Dependencies

Signed provider data contract
Verified workspace entitlement

Documentation Contact support

Tell us what you want answered, and we will tell you what it takes.

Write to us about Distribution Intelligence See the other capabilities