Use case · restaurants & retail

The daily close where every fee ties back to the settlement

The agent reads the files your operation already exports — POS reports, acquirer settlements, supplier invoices — and writes a close where every sale, fee and settlement points to its source line. Each manager sees only their own outlet.

The question this use case answers

How does a Singapore multi-outlet operator reconcile POS takings with acquirer settlements and compare margin per outlet with every figure traceable to its source report?

The files you already have

  • POS end-of-day reports (.xlsx / CSV)
  • Acquirer / NETS settlement statements
  • Supplier invoices (PDF / CSV)
  • Bank statements (CSV / MT940)
sample

What was the effective acquirer fee in May?

2.7%

settlement_may_outletA.csv · D2:D310 · you have access

Illustrative figure (sample) with fictional data — not a customer result or benchmark.

How the agent handles it, step by step

  1. 1

    Upload the day's or month's close

    Drop in the POS reports and acquirer settlements. No integration — the agent identifies each format on its own.

  2. 2

    The agent reconciles sale, fee and settlement

    Every figure carries where it came from: file, tab and line. The effective-rate and margin math is deterministic — the AI writes the prose, never the number.

  3. 3

    A settlement with no statement becomes a gap

    If an acquirer statement hasn't arrived, the agent flags the gap instead of closing the till over a hole.

  4. 4

    Schedule the per-outlet close

    Save the recipe and the agent re-runs it each period, comparing outlets with the same versioned audit trail.

Permission at synthesis time

Across a chain, each manager answers for their own outlet — and should not see the neighbouring one's result. Redijo checks the permission of each fragment at synthesis time: a manager sees their own outlet's margin, and another outlet's figure never enters the answer, not even inside a chain total they cannot open.

In F&B and retail, margin lives in the details: the acquirer fee that crept up, the settlement that came in light, the purchase booked at a different price. The trouble is those figures are scattered across POS reports, acquirer statements and invoices — and nobody can trace the result back to the source line. Generic AI confidently summarises it and hides exactly where the money leaked.

Redijo starts with the proof. The agent reads the files your operation already exports and writes a close where every sale, fee and settlement points to its source line.

Auditable by construction

The proof is born with the answer. Before writing a figure, the agent locates it in a file you uploaded and records its origin — file, tab, line. The effective-rate and margin math is deterministic; the AI writes the prose around it, never the number. That is why the close is auditable, not just plausible.

Each manager sees only their own outlet

The second win shows up when you compare outlets. Almost every tool checks access only at retrieval — then lets the AI blend everything into one answer. Redijo checks the permission of each fragment at synthesis time: a manager sees their own outlet’s margin, and another outlet’s figure never enters the answer, not even hidden inside a chain total.

Who it’s for

For the chain owner comparing outlets, the manager closing the day’s till, and the controller who needs every fee traced back to the statement. Your data is hosted in the cloud, with international-transfer safeguards.

Questions this page answers

Each outlet's data is hosted in the cloud, with international-transfer safeguards. PDPA-aligned by architecture — never by local servers.

Close the till with figures you can audit

Get started and reconcile POS and acquirers from the files your operation already exports.

PDPA-ready · DPA for businesses · cancel anytime

← Back to use cases by sector