Checked instantly on submission
The system checks that guest's submission frequency, receipt age, and receipt value before any points are awarded.
Features
Control layers keep self-serve points earning from becoming a loophole: hourly and daily receipt limits, receipt age, value thresholds, brand-wide image matching, and risk-tiered actions.

Updated August 2026 — This is the first question every venue owner asks, and the answer isn't 'trust the guest.' Every receipt submission runs through multiple control layers: hourly and daily submission limits, receipt age, value thresholds, image matching against every receipt already submitted across the brand, and automatic action based on risk level.
Updated: 2026-08-09
The system checks that guest's submission frequency, receipt age, and receipt value before any points are awarded.
New images are compared against receipts submitted in the last 90 days across every branch. A match flags both submissions, not just the later one.
Critical risk is blocked instantly. High risk is held for review. Medium risk gets points awarded but flags an admin to check.
| Control | What it catches | Setting |
|---|---|---|
| Rate limits per hour and per day | A single member can only submit so many receipts in an hour, and so many in a day. This caps how fast any one account can claim, whatever else it tries. | Suggested start: 3 per hour, 10 per day |
| Receipt age | A bill older than your limit is rejected, so nobody stockpiles receipts and claims them in a batch later. | Configurable per venue |
| Value thresholds, upper and lower | Bills below your floor are ignored so trivial claims do not farm points, and bills above your ceiling are held for a human look. | Configurable per venue |
| Duplicate image matching across 90 days, brand-wide | Every submitted photo is matched against submissions from the whole brand for the last 90 days. When the same image turns up twice, both submitters are flagged — not just the second one. | Similarity threshold configurable per venue |
| Repeated failed-scan detection | A member who keeps sending images the system cannot read is usually probing to learn what gets accepted. That pattern is flagged on its own. | Always on |
| Risk-tiered actions | Severe risk is blocked outright. High risk is held pending approval. Medium risk goes through but notifies an administrator, so you see the pattern without punishing the customer. | Always on |
The image is matched against every receipt the brand has received in the last 90 days. A match flags both submissions and routes them to review instead of awarding points.
The suggested starting default is 3 receipts per hour and 10 per day per guest. Your venue can adjust it to match its real traffic.
There are configurable receipt-age and value thresholds, upper and lower. Receipts outside those thresholds are rejected or routed to review, depending on setup.
The system tracks each guest's chain of failed scans. An abnormal streak is flagged as probing behavior and that account moves to a higher risk tier.
The system sorts submissions by risk tier, but your venue keeps final say. Critical risk is blocked automatically; High sits in a review queue; Medium gets points awarded with an admin notification.
PEKO and LOOP are two products from the same company. PEKO is the AI retention layer and runs alongside the POS you already use. LOOP is the AI-native POS that covers operations: recipe-level inventory, staff shifts, table plans and the kitchen display. Each works on its own, and run together they share one dataset, so nothing has to be entered twice. See PEKO + LOOP in one ecosystem