UAE · ENBook a demo
Procure AIProductInvestigator

Platform Analytics Agent

Tracks platform funnels and client user retention and explains drops.

Platform Analytics AgentProduct Analytics Agent · Strategy and Roadmap
Industry
Procure AI
Department
Product
Team
Strategy and Roadmap
Kind
Investigator
Approver
Head of Product
Code
PRD-06

It tracks the funnels that make our platform work for clients, such as requisition to PO, PO to goods receipt to matched invoice, and supplier invitation to supplier onboarded, and the retention of client users. When a step drops or a cohort retains worse it tests competing causes: a release, a client configuration change, a new supplier mix, an e-invoicing format change, a tracking change or month-end seasonality. **The problem it solves:** Why a platform funnel step, such as first-time-match or supplier onboarding, dropped or a client cohort retained worse: which step, from when, for which clients, and whether a release, a configuration, the supplier mix, tracking or seasonality caused it.

How it works

6 steps, and a person signs off before step 6.

  1. 01Looks it up

    Finds the record the request is about in the spend analytics warehouse: platform events by funnel step, client and week, with release and configuration change logs..

  2. 02Reads the request

    Works out what is being asked and what a complete answer needs.

  3. 03Checks the facts

    Checks figures, dates and rules against the source records, not from memory.

  4. 04Works it through

    Weighs competing explanations against the evidence and shows which one holds up.

  5. 05Asks a person

    Hands the work to Head of Product with its reasoning before anything goes out.

  6. 06Acts

    Once approved, it acts in the roadmap tool: it files an insight with its owner and the next step. and records what was done.

What it works with

Your knowledge, your systems, your people.

Knows

The platform funnel definitions, event tracking plan, release history, client configuration changes, known tracking gaps and the seasonal patterns of month-end and year-end buying.

Reads from

the spend analytics warehouse: platform events by funnel step, client and week, with release and configuration change logs.

Acts in

the roadmap tool: it files an insight with its owner and the next step.

Hands back

An object with funnel, period, client, change_percent, causes (a list with cause, evidence, confidence and how_to_confirm), top_cause and next_step.

Asks a person when

The diagnosis attributes the change to a named release, team or client configuration, or the top cause rests on data with a known tracking gap.

ApproverHead of Product

Settings

Your limits, set by your managers.

SettingDefaultWhat it controls
Change that counts (%)10A step moving more than this against baseline is investigated.
Baseline period (weeks)6The baseline the change is measured against.
Smallest cohort100Cohorts smaller than this are not read.
Rule out tracking changes firstOnWhen on, tracking changes are checked before any product cause.
Adjust for month-end and year-endOnWhen on, the baseline is matched to the same point in the period cycle.

Limits are settings, not code: a manager changes them without a developer.

Rules it always follows

  • A step moving more than 10 percent against the 6-week baseline is investigated.
  • When on is true, tracking and data-pipeline changes are ruled out first.
  • When on is true, month-end and year-end peaks are compared like for like.
  • A client configuration change, such as a new approval step or tolerance, is tested as a cause before a release.
  • Cohorts under 100 users are not read; every cause is ranked by evidence with what would confirm it.

See the Platform Analytics Agent agent on your own work.

We’ll run it on a sample of your work and show you what your approver would see.

Platform Analytics Agent · Procure AI | Basinwright Industries