GROC Fraud SOP

Overview

The purpose of this SOP is to prevent fraudulent orders that have been placed using stolen/cloned cards from being delivered to the fraudulent customer profile and to blacklist the profile that has placed the order. This SOP is also to prevent FTB Coupon Abuse. 

1.PNP Groceries Fraud Prevention Process

There are 2 different scenarios: 

Hoax FTB Coupon Abuse - Customers are creating new profiles to 

High Value Liquor Orders 

1.1.Hoax FTB Coupon Abuse

1.	Order appears on “GROC:Hoax FTB Orders” dashboard This dashboard was created by Ops and ringfences customers who repeatedly place orders with an FTB coupon in the same location with multiple newly created profiles with or without the same name and surname. Refer to the dashboard below to see the FTB coupon redemptions in the same location. Order Status is:  Picking_instruction_received  Picking_in_progress Delivery_instruction_received En_route_to_collection At_store If order has been collected:  Do not cancel, leave a note on branch dashboard  “Please do not hand over order as this is a potential hoax order”   Driver to fail the order and return to store
2.	Investigate Customer  2.1 Check Customer Profile  Was the profile created in the last 2 days?* Is the order value kept to the MOV (Minimum Order Value) required to redeem the coupon on all profiles?* Does the customer name have multiple profiles with different email addresses? Do these email addresses look similar? Are the email addresses sequential or have repetitive numbers? E.g. [e-mail] and [e-mail] Do most of these profiles have 1 grocery order? Are the addresses on the multiple profiles the same or similar?
3.	Decide if order should be canceled  3.1 Cancel the order IF Based off the checks you completed you have answered Yes to 1 or more FTB (*) questions under “Check Customer Profile” Cancel the order using the reason “Potential Hoax/Fraud” Leave a comment  “FTB Coupon Abuse”  Leave a note on the branch dashboard informing TDT  “Please note this order has been canceled due to a potential hoax” 3.2 Proceed with Order IF: Based off the checks you completed you have answered No to both FTB (*) questions under “Check Customer Profile” Customers profile was not created in the last 2 days  Customer did not order for the minimum order value  Leave a note on Admin Customer profile checked and proceed with order OR  Escalated to TL and decided to proceed with order  NB: If the review of the profile indicates that the order should proceed based on the above criteria however you identify other factors that indicate the order should be canceled consult TL before making a decision 3.3 Unsure whether to proceed with the order Inform TL and TL to decide what decision to take
4.	Create and Complete Exception Customer Resolution  Select “Other” Add a note “Potential hoax / fraud customer”  Apply a “full refund” to the customer  Select “add/Edit payout” Change Drop Down Option to “Yes” Store Resolution Mark the store resolution as “No Payment to the Store” Save Exception
5.	Blacklist Customer  Blacklist the customer using the following reasons  Reason: Coupon-Abuse Customer Leave a comment indicating “Customer was blacklisted due to meeting X/4 criteria required to blacklist them”

1.2. High Value Liquor Orders

1.	Order appears on “GROC: R4000 - R11000 Value Orders” dashboard Order Status is:  Picking_instruction_received  Picking_in_progress Delivery_instruction_received En_route_to_collection At_store If order has been collected:  Do not cancel, leave a note on branch dashboard  “Please do not hand over order as this is a potential hoax order”   Driver to fail the order and return to store
2.	Investigate Customer  2.1 Check Customer Profile  Was the profile created in the last 2 days? Does the customer name have multiple profiles with different email addresses? Do these email addresses look similar? Does the customer have multiple different cell phone numbers on each profile? Does the customer only have 1 GROC order? 2.2 Check Items Ordered Has the customer ordered multiple types of high value liquor? (eg 3 x Remy Cognac, 2 x Irish Whisky, etc) Has the customer ordered more than 3 or a large quantity of a specific liquor item? (e.g. 12 Smirnoff 1818 Round Vodka)  2.3. Check Order History Does the customer have a high proportion of abandoned orders relative to those delivered? Has the customer placed multiple liquor orders on the same day or in the previous days? 2.4 Check Banking Details on Orders Has the customer used multiple different bank cards on different orders OR Different card tried on the same order and/or Bank Response on an order is “Declined Call for approval” or “Declines Card Reported as Stolen” or "Declined Auth Declined"
3.	Decide if order should be canceled  3.1 Cancel the order IF Based off the checks you completed you have answered Yes in 3 or more sections  Cancel the order using the reason “Potential Hoax/Fraud” Leave a comment “Multiple High Value Liquor Orders/High Value Liquor Order” Leave a note on the branch dashboard informing TDT  “Please note this order has been canceled due to a potential hoax” 3.2 Proceed with Order IF: Customer has a lot of orders but all with the same card  Order is only groceries and not liquor Leave a note on Admin Customer profile checked and proceed with order OR  Escalated to TL and decided to proceed with order  NB: If the review of the profile indicates that the order should proceed based on the above criteria however you identify other factors that indicate the order should be canceled consult TL before making a decision 3.3 Unsure whether to proceed with the order Inform TL and TL to decide what decision to take
4.	Blacklist Customer  Blacklist the customer using the following reasons  Reason: Credit Card Fraud Customer Leave a comment indicating “ “Customer was blacklisted due to meeting X/4 criteria required to blacklist them”
5.	Send Macro to Finance Send Customer Credit Card Fraud - Blacklisting macro to Finance

2. Examples of Products with High Risk to Credit Card Fraud

• Vodka
• Whisky 
• Cognac

3. Decision Tree

4. Monitoring Dashboards

 GROC: Hoax FTB Orders

GROC: R4000 - R11000 Value Orders

5. Labeling a Customer as a Complainer

The purpose of this section is to provide a guide to Agents on how to identify customers who are abusing the compensation process, define the procedure for tagging them as complainers and provide a guide on the resolution for customers with the complainer tag. Complainers should not be blacklisted. The compensation should just be rejected.

5.1. Profile Tagging 

Criteria to mark a customer as a Complainer:

1.	First Criteria  The Customer's Total Spend is lower than their Total Basket spend e.g:
2.	Second Criteria Agents can use the customer’s Email on Zendesk to find all complaints linked to the customer.  The Customer has 2 or more consecutive complaints on orders where images are required and not received:  The Customer complained about an incorrect item received without providing images and compensation was given  The Customer complained about a visible PFQ issue without providing images and compensation was given  The Customer has complained about mishandled items without providing images and compensation was given. OR The Customer has 2 or more consecutive complaints on the following:  The Customer has complained about missing items and compensation was given. The Customer has complained about an invisible PFQ issue and compensation was given. The Customer has complained about a Marked Delivered not Received where the delivered timestamp was right outside the Customers address. OR The Customer has received compensation for 2 or more of the same complaint type in the last 30 days. OR Customer has multiple profiles with similar complaints  (Check Name, Email or Cell Phone Number)  OR Percentage of orders with manual exceptions > 30% If between 15% - 30% - TL to Review
3.	Labeling Customer on Admin Go to the Customers profile on Admin  Select ‘Actions’  Select ‘Edit Customer Info’ Select ‘Treatment Status’ Select ‘Complainer’ Select Save
4.	Labeling Customer on Zendesk  Click on the Customers name on the Zendesk Ticket  Scroll down on the left panel until you reach ‘Mr D Food Customer Classification’ Select Demanding once identified that the Customer fits the criteria to be labeled as a Complainer

5.2. Resolving Customers with a Complainer/Demanding Tag 

ALL compensation that is given to a Customer with a Complainer/Demanding tag needs to be approved by the MOD on shift. 

Do NOT blacklist these Customers, only refuse compensation. FTB Coupon Abuse customers must still be blacklisted.

When refunding a complainer, agent is to add a note in the exception: Authorized by MOD: MOD Name (Complaint Type) 

6. Deactivation Reasons for Blacklisting

Deactivation Reason	Scenario	Customer Examples
Deactivation Coupon Abuser	Multiple profiles being created to redeem FTB or existing customer marketing coupons (Not CS coupons. CS coupon abuse is complainer behaviour. These customers should not be blacklisted. Only withhold compensation) Reason to be used when customer order comes up on GROC: FTB Coupon Abuse Orders Kibana Dashboard	12035431 12065710 12089896
Deactivation Credit Card Fraudster	Suspicious customer making use of multiple credit cards with transaction logs showing responses like "Authentication Declined" or "Blocked by risk system" Reason to be used when a customer order comes up on GROC: R3500 - R11000 Value Orders Kibana Dashboard and it has been deemed as fraud	12095733 12432005 12394342
Deactivation Hoax Order Customer	Hoax customer placing an order to lure the driver or a Customer finds a loophole in the app and manipulates it to get free Groceries. E.g. Customer put negative tip to get free groceries (Issue has been fixed)	8473166
Deactivation Offensive Customer	Customer showing offensive behaviour to our drivers or CS team
Deactivation Out of Area Customer	Potential customer that is outside of our polygon pinning themselves on the edge of the nearest polygon, but more than 500m from that location. Not a coupon abuser moving their pin.
Deactivation Requested By Customer
Deactivation Suspension Under Investigation
Deactivation Violent Customer	Customer showing violent behaviour to our drivers or CS team
Fraudulent account access	For account takeovers