# RevCent Customer Overview This document gives a broad overview of how customers work in RevCent. The customer aspect of RevCent should be understood as a customer intelligence layer for ecommerce businesses. It connects identity, purchases, subscriptions, payments, support, metadata, groups, engagement, reporting, and automation into a single customer-centered view. This explains why storing customers inside RevCent is powerful, what kinds of customer data and relationships RevCent can hold, and how customer data unlocks ecommerce capabilities across reporting, automation, AI, subscriptions, payments, engagement, customer portals, customer service, and personalization. --- ## What Customers Represent in RevCent The customer aspect of RevCent is the central place where RevCent stores customer identity, contact data, purchase history, payment relationships, metadata, groups, notes, and related ecommerce activity. A customer in RevCent is not just a name and email address. A customer can become the center of a full ecommerce profile that connects: - Contact identity - Billing and address information - Internal customer IDs - Campaign association - Customer metadata - Customer groups - Customer payment methods - Sales - Product purchases - Shipments - Taxes - Discounts - Trials - Subscriptions - Subscription renewals - Chargebacks - Fraud detections - Notes - Lifetime value - Last purchase behavior - Engagement history - Support context - Automation and AI context When customer data and purchase behavior live in RevCent, the customer record becomes a powerful foundation for commerce operations. --- ## Why Storing Customers in RevCent Matters RevCent can automatically create a customer during an initial sale if the customer does not already exist. Customers can also be created before a purchase attempt, which is useful for prospects, leads, pre-checkout workflows, phone orders, AI workflows, manual sales processes, and third-party integrations. Once a customer is stored in RevCent, that customer becomes a long-term object that connects future activity. Instead of every purchase, support issue, subscription, or payment being isolated, RevCent can connect those events back to a customer identity. This enables: - Better customer history. - Better personalization. - Better segmentation. - Better reporting. - Better AI assistance. - Better support experiences. - Better payment recovery. - Better customer portals. - Better subscription management. - Better chargeback mitigation. - Better lifecycle automation. - Better understanding of customer value. --- # Customer Data Stored in RevCent A RevCent customer can include multiple layers of data. ## Identity and Contact Data Customer identity fields can include: - First name - Last name - Email - Phone - Company - Internal customer ID - Customer ID - Campaign association - Created date - Updated date - Enabled, disabled, or blocked status This gives RevCent a reliable customer identity object that can be reused across payments, subscriptions, AI tools, support, reporting, and automation. --- ## Address and Location Data Customer records can include: - Address line 1 - Address line 2 - City - State - ZIP / postal code - Country This can support: - Billing workflows - Shipping workflows - Customer verification - Fraud/risk analysis - Regional reporting - Tax-related context - Support interactions - AI Voice Agent verification and personalization --- ## Payment Method Context Customers can have associated customer cards. A stored customer card record can include safe card reference data such as: - Customer card ID - Card type - First 6 digits / BIN - Last 4 digits - Expiration date - Default payment method status - Created/updated timestamps This supports workflows such as: - Default payment method management - Subscription billing - Trial expiration billing - Pending sale recovery - Payment retry strategies - Customer support around payment methods - AI Voice Agent payment update flows - Customer portal payment management Important: full card handling should occur only through appropriate secure workflows. --- --- # Secure Customer Payment Data and PCI-Compliant Storage A major advantage of storing customers in RevCent is that customer payment data can be securely stored and associated with the customer for future payment activity. RevCent is PCI compliant and stores customer payment data in a separate PCI storage layer. Raw credit card data is not exposed to RevCent users, customer support teams, AI Assistants, AI Voice Agents, or general customer workflows. Instead, RevCent exposes only generic, safe payment-method reference data that is useful for customer, billing, and payment workflows. --- ## Separate PCI Storage Layer Customer payment data is handled through RevCent’s secure PCI storage layer. This separation is important because it allows RevCent to support payment capabilities without exposing raw cardholder data throughout the platform. Conceptually: ```text Customer record ↓ References customer payment method ↓ Secure PCI storage layer holds sensitive payment data ↓ RevCent workflows use secure payment reference when payment processing is needed ``` The customer record can be associated with payment methods, but the sensitive raw card data remains protected in the PCI storage layer. --- ## Generic Payment Method Data Exposed RevCent can expose generic payment-method reference data for customer and billing context. Examples of generic customer card data include: - Customer card ID - Card type - First 6 digits / BIN - Last 4 digits - Expiration date - Default payment method status - Created/updated timestamps This generic information helps identify which stored payment method is associated with the customer without exposing raw credit card data. Example: ```text Default card ending in 4242 Visa expiring 08/28 Customer card ID: XXXXXXXXXXXXXXXXXXXX ``` This is useful for customer context, billing workflows, subscription management, and payment recovery while keeping sensitive payment data protected. --- ## Raw Credit Card Data Is Not Exposed Raw credit card data is not exposed to RevCent users, customer support teams, AI Assistants, AI Voice Agents, or general customer automation. The customer-facing and operational layers of RevCent may reference generic card details, but they do not expose the full underlying payment credentials. This means RevCent can support stored payment capabilities while keeping sensitive payment data isolated in the PCI-compliant payment layer. --- # How Secure Stored Payment Data Powers Ecommerce Because RevCent can securely associate payment methods with customers, businesses can support important ecommerce workflows that depend on stored customer payment data. --- ## Subscription Renewals Stored customer payment data allows RevCent to process subscription renewals without requiring the customer to re-enter payment information for every renewal. Conceptual flow: ```text Customer has subscription ↓ Customer has stored payment method ↓ Subscription renewal becomes due ↓ RevCent uses secure payment reference ↓ Renewal payment is attempted ``` This supports recurring revenue businesses by making renewal billing possible through securely stored customer payment methods. --- ## Trial Expiration Billing Stored customer payment data can also support trial expiration billing. Example: ```text Customer starts trial ↓ Payment method is securely stored ↓ Trial expires ↓ RevCent attempts payment using secure stored payment reference ``` This helps businesses convert trials into paid subscriptions or paid purchases without requiring a manual payment re-entry step at the moment of trial expiration. --- ## Decline Salvage and Payment Recovery Securely stored customer payment data also enables decline salvage and payment recovery workflows. Decline salvage means attempting to recover revenue from failed or declined payment attempts. Examples: - A subscription renewal declines. - A trial expiration payment declines. - A pending sale needs recovery. - A stored card may need to be retried according to business rules. - A customer may need to update their payment method. - A recovery workflow may send email or trigger AI Voice Agent outreach. Because the customer has a stored payment relationship, RevCent can support recovery workflows around that customer and payment method. Conceptual flow: ```text Payment attempt declines ↓ Customer and stored payment method remain associated ↓ RevCent can support retry, recovery, or update workflows ↓ Declined revenue may be salvaged ``` This is especially important for ecommerce businesses with subscriptions, trials, recurring billing, or payment recovery processes. --- ## Customer Lifetime Value and Stored Payment Data Stored payment methods also support more accurate customer value over time. When RevCent can securely process renewals, trials, and recovered payments, the customer record can accumulate richer lifetime value data. This can include: - Initial sale value - Subscription renewal value - Remaining amount - Amount to salvage - Refunded amount - Net value - Gross value - Renewal counts - Overdue renewal counts - Last sale date - Last subscription renewal date This makes stored customer payment data part of the broader customer intelligence layer. --- ## Why This Matters Secure customer payment storage gives RevCent customers a powerful foundation for ecommerce growth. It enables: - Recurring billing - Subscription renewals - Trial expiration billing - Payment retry strategies - Decline salvage - Pending sale recovery - Customer portal payment method management - Payment update workflows - Revenue recovery - More complete lifetime value tracking The key concept is: ```text RevCent securely stores customer payment data in a separate PCI storage layer, exposes only generic payment-method reference data, and uses the secure stored payment relationship to support renewals, trial billing, and decline salvage. ``` ## Metadata Customer metadata is one of the most powerful customization tools. Metadata is a flexible name/value system that allows a business to attach custom data to customers. Examples: ```text affiliate_id = aff_123 lead_source = google_ads customer_tier = vip sales_rep = sarah preferred_language = en crm_id = 987654 loyalty_status = gold risk_flag = manual_review marketing_segment = reorder_candidate ``` Metadata helps RevCent adapt to each business model. Because every business thinks differently, metadata lets a customer record store business-specific meaning that may not fit into standard fields. Metadata can power: - Reporting - BigQuery analysis - Customer segmentation - AI Assistant context - AI Voice Agent personalization - Email personalization - Function logic - Customer group logic - Payment Profile filtering - Customer support routing - Marketing workflows - Reorder campaigns - Winback campaigns Important concept: Users may describe metadata without saying “metadata.” For example: ```text Group customers by affiliate. ``` This may actually mean: ```text Group customers by metadata where name = affiliate_id. ``` RevCent users and automation workflows should recognize that custom business labels are often stored as metadata. --- # Customer Groups Customer Groups allow businesses to organize customers into named segments. Customer groups can represent: - VIP customers - Wholesale buyers - Subscription customers - Trial customers - At-risk customers - High lifetime value customers - Refund-prone customers - Fraud review customers - Affiliate-driven customers - Campaign audiences - Geographic segments - Loyalty tiers - Sales-rep-owned accounts - Support escalation groups Customer Groups are useful because they create simple, reusable segments that can be referenced across workflows. --- ## Why Customer Groups Are Useful Customer Groups help businesses move from individual customer records to organized customer strategy. They can support: - Segment-based email follow-up - AI Assistant targeting - AI Voice Agent targeting - Customer portal personalization - Payment Profile filtering - Subscription retention workflows - VIP service workflows - Fraud/risk review - Manual support prioritization - BigQuery reporting - Campaign performance analysis Example: ```text Customers in "VIP" group can receive priority support and more personalized AI Voice Agent handling. ``` Example: ```text Customers in "Subscription Save" group can be contacted by an AI Voice Agent before cancellation. ``` --- --- # Customer Groups as Granular Segmentation Engines Customer Groups are one of the most powerful parts of the customer aspect of RevCent because they allow customers to be organized on a highly granular basis. A Customer Group is not only a simple label like: ```text VIP ``` or: ```text Wholesale ``` It can also represent a dynamic business rule. Customer Groups can be used to identify customers based on detailed qualifiers such as: - Campaign - Product group purchased - Third-party shop - Metadata - Account age - Lifetime value - Lifetime refunded amount - Number of sales - Average sale amount - Days since last sale - Successful sale count - Upsell sale count - Declined sale count - Abandoned sale count - Fraud detection count - Lifetime chargeback amount - Chargeback count - PayPal dispute amount - PayPal dispute count - Subscription renewal history - Overdue renewal count - Subscription status - Membership in other customer groups This makes Customer Groups a powerful way to turn raw customer behavior into meaningful business segments. --- ## Qualifier Methods Customer Groups can be structured in different ways depending on how customers should qualify. ### No Qualifiers A group can have no qualifiers. This means customers are added or removed manually or by a workflow. Examples: ```text Manual VIP Review Needs Human Support Approved For Special Offer Do Not Contact ``` This is useful when group membership is based on human judgment, AI classification, or an external system. --- ### Specific Value Qualifiers A group can qualify customers based on specific values and filters. This is where Customer Groups become highly granular. Examples: ```text Customers with lifetime value greater than $500 Customers with at least 3 successful sales Customers with no sales after account creation Customers with more than 1 declined sale Customers who purchased from a specific product group Customers from a specific campaign Customers with metadata affiliate_id = aff_123 Customers with subscription status = overdue Customers with at least 2 overdue renewals Customers with chargeback count greater than 0 ``` These qualifiers allow RevCent to create segments from actual behavior and stored customer data. --- ### Customer Group-Based Qualifiers A group can also be based on membership in other customer groups. This allows layered segmentation. Examples: ```text Customer must be in VIP group Customer must be in Subscription Customer group Customer must not be in Do Not Contact group Customer must be in both High LTV and Active Subscriber Customer must be in any of several campaign groups ``` This enables composite customer groups. Example: ```text VIP Winback Candidates = In VIP group + Not in Active Subscriber group + Days since last sale > 90 ``` --- # Examples of Granular Customer Groups ## VIP Customers ```text Lifetime value >= $1,000 AND chargeback count = 0 AND customer is not in Manual Review group ``` Use cases: - Priority support - VIP email campaigns - AI Voice Agent follow-up - Special offers - Human escalation --- ## Winback Candidates ```text Lifetime value >= $250 AND days since last sale >= 90 AND not in Do Not Contact group ``` Use cases: - Winback email - AI Assistant-generated offer - AI Voice Agent follow-up - Customer success review --- ## Failed Payment Recovery Group ```text Declined sale count >= 1 OR overdue renewal count >= 1 ``` Use cases: - Payment recovery email - AI Voice Agent payment update call - Function-triggered CRM task - Internal support queue --- ## High Refund Risk Group ```text Lifetime refunded amount >= defined threshold OR refunded amount is high relative to lifetime value ``` Use cases: - Manual review - Support escalation - Fraud/risk review - Modified offer eligibility --- ## Subscription Save Group ```text Subscription status = cancelled or overdue AND lifetime value >= defined threshold ``` Use cases: - Retention email - AI Voice Agent save call - AI Assistant churn analysis - Human retention review --- ## Prospect Group ```text Customer has no sales AND days since created >= defined threshold ``` Use cases: - Lead nurture - First-purchase offer - AI Assistant qualification - Email sequence --- ## Affiliate Segment ```text Metadata name = affiliate_id AND metadata value = specific affiliate ``` Use cases: - Affiliate reporting - Affiliate-specific follow-up - Performance analysis - Customer quality analysis --- # Customer Groups as Workflow Triggers The most powerful use of Customer Groups is not only segmentation. It is what happens when group membership changes. When a customer is added to or removed from a group, RevCent can use those changes as meaningful customer lifecycle events. Important customer group events include: ```text customer.updated.customer_group.added customer.updated.customer_group.removed ``` These events can signal that a customer has entered or exited a business-defined state. Examples: ```text Customer enters VIP group Customer leaves Trial Customer group Customer enters Failed Payment Recovery group Customer leaves Winback Candidate group Customer enters Manual Review group Customer leaves Subscription Save group ``` Each of these events can trigger additional automation. --- ## Why Group Added / Removed Events Matter A customer being added to a group is often a business signal. Examples: ```text Customer added to VIP = customer is valuable enough for special handling ``` ```text Customer added to Failed Payment Recovery = customer may need payment recovery outreach ``` ```text Customer added to Fraud Review = customer may need manual review before further action ``` ```text Customer removed from Winback Candidate = customer may have purchased again or no longer needs winback outreach ``` Group changes can represent milestones, risk changes, lifecycle changes, or engagement opportunities. This makes Customer Groups useful not only for reporting, but also for automation. --- # Automations Triggered by Customer Group Changes When `customer.updated.customer_group.added` or `customer.updated.customer_group.removed` occurs, RevCent can trigger a wide range of workflows. These can include: - Functions - AI Assistants - AI Voice Agents - Email Templates - External endpoints through Functions - Third-party AI agents - CRM updates - Helpdesk tickets - Internal alerts - Metadata updates - Notes - BigQuery/reporting workflows - Customer success workflows The group change becomes the starting point for action. Conceptual flow: ```text Customer qualifies for group ↓ Customer group membership changes ↓ customer.updated.customer_group.added or customer.updated.customer_group.removed event occurs ↓ Automation runs ↓ Customer receives outreach or internal workflow is triggered ↓ Outcome is stored back on the customer through notes, metadata, or group changes ``` --- ## Functions Triggered by Group Changes Functions can run custom code when customer group membership changes. Examples: ```text Customer added to VIP group ↓ Function sends Slack alert to customer success ``` ```text Customer added to Failed Payment Recovery group ↓ Function creates CRM task ``` ```text Customer removed from Do Not Contact group ↓ Function re-enables eligible outreach workflow ``` ```text Customer added to Fraud Review group ↓ Function sends customer data to external risk system ``` Functions are especially useful when group changes need to connect RevCent to third-party systems. Examples of third-party destinations: - CRM - Helpdesk - Slack / Teams - Marketing platform - Customer success platform - Internal operations system - Data warehouse - External AI Agent - Fraud/risk platform - Fulfillment or logistics platform --- ## AI Assistants Triggered by Group Changes AI Assistants can be triggered when a customer enters or exits a group. This is useful when the group change requires reasoning, summarization, or a next-best-action decision. Examples: ```text Customer added to High LTV At Risk group ↓ AI Assistant summarizes customer history and recommends retention action ``` ```text Customer added to Refund Risk group ↓ AI Assistant reviews purchase/refund history and creates internal memo ``` ```text Customer added to Winback Candidate group ↓ AI Assistant recommends offer, message, and channel ``` ```text Customer removed from Failed Payment Recovery group ↓ AI Assistant determines whether recovery succeeded and updates summary ``` AI Assistants can help transform group membership changes into intelligent customer decisions. --- ## AI Voice Agents Triggered by Group Changes Customer group changes can also trigger AI Voice Agent workflows. Examples: ```text Customer added to Failed Payment Recovery group ↓ AI Voice Agent calls customer to help resolve payment issue ``` ```text Customer added to Subscription Save group ↓ AI Voice Agent calls customer with retention-focused instructions ``` ```text Customer added to VIP Follow-Up group ↓ AI Voice Agent places personalized follow-up call ``` ```text Customer removed from Trial Candidate group and added to Trial Expired group ↓ AI Voice Agent calls customer about conversion options ``` AI Voice Agents are especially useful when the group change represents a time-sensitive customer moment. Examples: - Failed payment - Subscription cancellation risk - Trial expiration - High-value customer issue - Reorder opportunity - Customer success follow-up --- ## Email Templates Triggered by Group Changes Group changes can also power email workflows. Examples: ```text Customer added to Welcome Nurture group ↓ Send welcome email sequence ``` ```text Customer added to Winback Candidate group ↓ Send winback offer ``` ```text Customer added to VIP group ↓ Send VIP thank-you email ``` ```text Customer removed from Payment Recovery group ↓ Send payment resolved confirmation ``` Email is useful when the group change should result in branded, repeatable, customer-facing communication. --- ## Third-Party AI Agents and External Tools Customer group events can also be used by third-party AI agents and external tools. A third-party AI Agent can monitor or receive customer group events through RevCent-connected workflows and decide what to do next. Examples: ```text customer.updated.customer_group.added: VIP ↓ Third-party AI Agent enriches customer profile from CRM ↓ Agent recommends account manager follow-up ``` ```text customer.updated.customer_group.added: Winback Candidate ↓ Third-party AI Agent analyzes offer history ↓ Agent selects best incentive ↓ Email Template or external marketing platform sends outreach ``` ```text customer.updated.customer_group.added: Fraud Review ↓ External risk AI Agent reviews behavior ↓ Agent returns risk recommendation ↓ Customer metadata is updated ``` ```text customer.updated.customer_group.removed: Failed Payment Recovery ↓ External AI Agent marks recovery complete ↓ CRM task is closed ``` This makes Customer Groups a bridge between RevCent and broader AI-driven business systems. --- # Group Membership as a Customer Lifecycle State Customer Groups can represent lifecycle states. Examples: ```text Prospect First-Time Buyer Repeat Buyer VIP At Risk Winback Candidate Subscription Active Subscription Overdue Trial User Trial Expired Payment Recovery Do Not Contact Manual Review Fraud Review Chargeback Risk ``` When a customer moves between groups, that movement can represent a lifecycle transition. Example lifecycle: ```text Prospect ↓ First-Time Buyer ↓ Repeat Buyer ↓ VIP ``` Another lifecycle: ```text Subscription Active ↓ Subscription Overdue ↓ Payment Recovery ↓ Recovered ``` Another lifecycle: ```text Trial User ↓ Trial Expiring Soon ↓ Trial Converted ``` or: ```text Trial User ↓ Trial Expired ↓ Winback Candidate ``` Each transition can trigger different engagement. --- # Group-Driven Customer Journeys Customer Groups can be used to design complete customer journeys. ## Payment Recovery Journey ```text Customer enters Failed Payment Recovery group ↓ Email Template sends payment update email ↓ If high lifetime value, AI Voice Agent calls customer ↓ Function creates CRM task ↓ Outcome stored as note or metadata ↓ Customer removed from group when recovered ``` ## VIP Journey ```text Customer enters VIP group ↓ AI Assistant summarizes customer history ↓ Customer success team receives notification ↓ VIP email is sent ↓ Support rules prioritize customer ``` ## Winback Journey ```text Customer enters Winback Candidate group ↓ BigQuery or AI Assistant evaluates purchase history ↓ Offer is selected ↓ Email Template sends offer ↓ AI Voice Agent follows up if customer is high value ``` ## Risk Review Journey ```text Customer enters Risk Review group ↓ Function sends data to risk system ↓ AI Assistant summarizes customer history ↓ Human reviews before additional outreach ``` --- # Guardrails for Group-Triggered Automation Because customer group changes can trigger powerful automation, group-driven workflows should be designed carefully. Recommended guardrails: - Clearly define what each group means. - Use descriptions that explain the group’s purpose and qualifiers. - Avoid vague group names. - Avoid triggering customer outreach from ambiguous groups. - Add cooldown rules for repeated outreach. - Respect do-not-contact or suppression groups. - Ensure AI Voice Agent calls are appropriate for the customer state. - Make sure Functions and external tools record outcomes. - Use notes and metadata to preserve automation history. - Use human review for sensitive groups such as fraud, chargeback, legal, or angry customers. - Avoid circular workflows where group changes repeatedly trigger each other. - Test group qualifiers before using them for live outreach. - Disable workflows until group behavior is verified. --- # Why Customer Groups Are So Powerful Customer Groups combine three important capabilities: ```text Granular qualification + lifecycle state tracking + event-triggered automation ``` That means a customer group can do much more than label a customer. It can identify a precise customer state and trigger the right next action. Examples: ```text High-LTV customer becomes inactive ↓ Winback journey starts ``` ```text Customer has overdue renewal ↓ Payment recovery workflow starts ``` ```text Customer becomes VIP ↓ Customer success workflow starts ``` ```text Customer enters fraud review ↓ Risk workflow starts ``` This makes Customer Groups one of the most important building blocks for automated, intelligent ecommerce operations in RevCent. # Purchase and Behavior Data One of the biggest advantages of storing customers in RevCent is that purchase behavior can be connected back to the customer. A customer can be associated with: - Sales - Product sales - Products purchased - Shipping records - Taxes - Discounts - Trials - Subscriptions - Subscription renewals - Chargebacks - Fraud detections - Notes - Lifetime value - Last sale date - Last subscription renewal date This turns the customer record into an ecommerce behavior profile. --- ## Products Purchased Customer records can include products purchased over the customer’s lifetime. This can include details such as: - Product ID - Product name - Internal product ID - SKU - Quantity purchased - Average purchase price - Gross amount - Refunded amount - Purchase count - Purchase dates This enables questions like: ```text What products has this customer bought? Which customers bought Product A but not Product B? Who should be targeted for a replenishment campaign? Which customers are likely good candidates for an upsell? ``` --- ## Lifetime Value Customer records can include lifetime value data. This can include: - Gross amount - Net amount - Refunded amount - Remaining amount - Amount to salvage - Total amount - Discounted amount - Number of sales - Average sale amount - Subscription renewal totals - Overdue renewal counts - Chargeback totals - Fraud detection counts Lifetime value allows businesses to treat customers differently based on actual economic value. Examples: - VIP customers can receive higher-touch support. - High-LTV customers can get retention offers. - Subscription renewal failures can be prioritized by customer value. - Chargeback risk can be weighed against purchase history. - AI Voice Agents can tailor call urgency or escalation based on value. --- # Status, Blocking, and Notes Customers can have statuses such as enabled, disabled, or blocked. This matters for: - Risk management - Support workflows - Payment attempts - Customer access - Customer portal behavior - Subscription handling - Fraud prevention - Manual review Customer notes allow human context to be attached to the customer. Notes may include: - Support summaries - Call outcomes - Refund reasons - Shipping issues - Escalation notes - Fraud review notes - AI Voice Agent call summaries - Payment recovery - PCI-compliant customer payment data storage in a separate secure payment layer, supporting subscription renewals, trial billing, and decline salvage outcomes - Manual review decisions - Customer preference notes Notes are useful because not every important customer fact is structured. --- --- # Notes as Customer Memory and Automation Triggers Notes are one of the most important ways RevCent can preserve human, AI, support, and operational context. A note can be attached directly to a customer, but notes can also be created for other customer-related records such as: - Customer - Sale - Product sale - Shipping - Subscription - Subscription renewal This matters because a note can capture context wherever the customer interaction happened, while still contributing to the broader customer history. Examples: ```text Customer note: Customer requested callback tomorrow. Sale note: Customer said charge was authorized after support call. Shipping note: Customer reported package not received. Subscription note: Customer wants to pause subscription after current cycle. Subscription renewal note: Renewal failed; customer said they will update card Friday. ``` Notes are useful because not every important customer detail fits neatly into structured fields. --- ## Notes Attached to Customers Customer notes can act as a support memory layer. They can record: - Support call summaries - AI Voice Agent call outcomes - Refund promises - Cancellation requests - Escalation instructions - Customer preferences - Payment recovery - PCI-compliant customer payment data storage in a separate secure payment layer, supporting subscription renewals, trial billing, and decline salvage attempts - Customer objections - Risk review decisions - Shipping issue summaries - Internal team comments - Follow-up commitments - Human override decisions This helps every future support interaction start with more context. Example: ```text Customer called about delayed shipment. Explained that replacement will be sent if tracking does not update within 48 hours. Customer was calm and prefers SMS follow-up. ``` This note can help: - Human support agents - AI Assistants - AI Voice Agents - Functions - Customer success workflows - Reporting and audit workflows --- ## Notes Attached to Related Customer Records A customer’s history is not only stored on the customer object itself. Important context may be attached to related records. Examples: | Note Location | Example Use | |---|---| | `customer` | General customer support context. | | `sale` | Charge, refund, order, or purchase-specific context. | | `product_sale` | Product-specific issue, replacement, or warranty context. | | `shipping` | Delivery issue, tracking problem, package not received, address correction. | | `subscription` | Cancellation, pause, retention, customer preference. | | `subscription_renewal` | Renewal failure, recovery attempt, customer payment promise. | This allows RevCent to preserve context at the correct level. A customer-level note is good for general history. A sale-level note is better when the note only applies to a specific sale. A shipping-level note is better when the note only applies to a specific shipment. --- # `note.created` as an Event Trigger A created note can be more than passive documentation. The `note.created` event can be used as a trigger for additional action. This means a note can become the starting point for automation. Conceptually: ```text Human, AI Voice Agent, AI Assistant, or Function creates a note ↓ note.created event occurs ↓ Function, AI Assistant, or external endpoint workflow runs ↓ Workflow reads note contents and related item context ↓ Workflow takes appropriate follow-up action ``` This is powerful because support notes often contain intent. Examples: ```text "Customer wants a callback tomorrow." ``` ```text "Customer is upset and requested a manager." ``` ```text "Customer said they will update payment on Friday." ``` ```text "Refund exception approved by supervisor." ``` ```text "Package not received; resend if no tracking update in 48 hours." ``` These notes can drive downstream action. --- ## Functions Triggered by `note.created` A RevCent Function can be configured to run when a note is created. The Function can inspect the note text, item type, item ID, customer context, metadata, or related record details and then perform custom logic. Examples: - Send a Slack alert when a note contains “manager requested.” - Create a ticket in a helpdesk when a support note contains “escalate.” - Call an external endpoint when a refund-related note is created. - Add metadata when a note indicates a customer requested cancellation. - Add the customer to a Customer Group when a note includes a retention status. - Notify a customer success team when a VIP customer has a negative support note. - Trigger a CRM update when a sales rep note is created. - Start a scheduled follow-up process when a note mentions a callback date. Example workflow: ```text note.created ↓ Function reads note text ↓ Function detects "callback tomorrow" ↓ Function creates task in external CRM or posts to internal endpoint ↓ Function inserts metadata: support_followup_required = true ``` Functions are useful because they can connect note events to almost any third-party system through outbound internet access and custom code. --- ## AI Assistants Triggered by `note.created` A `note.created` event can also be used to trigger an AI Assistant workflow. This is useful when the note needs interpretation, summarization, classification, or next-best-action reasoning. Examples: - Summarize a long support note. - Classify note sentiment as positive, neutral, negative, or urgent. - Determine whether the customer should be escalated. - Decide whether the customer belongs in a support group. - Generate an internal customer success memo. - Recommend whether email or phone follow-up is appropriate. - Identify whether a note suggests churn risk. - Identify whether a note suggests refund, chargeback, fraud, or legal risk. - Generate a follow-up email draft. - Decide whether an AI Voice Agent should call the customer. Example workflow: ```text Support agent creates note: "Customer is upset about second failed shipment and asked for supervisor." ↓ note.created triggers AI Assistant ↓ AI Assistant summarizes issue and marks escalation priority ↓ AI Assistant recommends human supervisor follow-up ↓ Metadata or note is added for support team visibility ``` This turns notes into intelligent triggers for customer support and customer success workflows. --- ## External Endpoint Workflows From `note.created` A `note.created` event can be used to send information to an external endpoint, typically through a Function. This allows RevCent notes to connect with outside systems. Examples of endpoints: - CRM endpoint - Helpdesk endpoint - Slack or Teams webhook - Internal operations API - Data warehouse ingestion endpoint - Marketing automation endpoint - Project management task endpoint - Customer success platform endpoint - Fraud/risk review system endpoint Example workflow: ```text note.created ↓ Function receives event ↓ Function formats payload ↓ Function POSTs to external endpoint ↓ External system creates ticket, task, alert, or record ``` Example payload concept: ```json { "event": "note.created", "item_type": "customer", "item_id": "XXXXXXXXXXXXXXXXXXXX", "note_text": "Customer requested cancellation save offer callback.", "recommended_action": "retention_followup" } ``` This lets notes become a bridge between RevCent customer context and the business’s broader operational systems. --- ## Note Contents Can Drive Conditional Automation The content of the note can determine what happens next. Examples: | Note Content | Possible Follow-Up | |---|---| | “callback” | Create CRM task or trigger reminder workflow. | | “refund approved” | Notify billing or support team. | | “manager” / “supervisor” | Escalate to human support. | | “angry” / “upset” | Mark customer for priority review. | | “cancel” | Trigger retention AI Assistant or add to Subscription Save group. | | “payment Friday” | Schedule payment recovery follow-up. | | “chargeback” | Notify risk team or create internal review. | | “package not received” | Trigger shipping support workflow. | | “VIP” | Notify customer success or route to high-priority queue. | RevCent workflows should treat notes as possible signals, not just plain text. --- ## Notes From AI Voice Agents AI Voice Agents should often create notes after calls. These notes can capture: - Call summary - Customer intent - Payment outcome - Subscription outcome - Refund/cancellation request - Escalation request - Customer sentiment - Whether the call reached the customer - Whether the customer requested follow-up - Whether the customer asked for a human A note from an AI Voice Agent can then trigger additional automation. Example: ```text AI Voice Agent creates note: "Customer answered. Payment failed because card expired. Customer requested payment update link by email." ↓ note.created triggers Function ↓ Function sends payment update email using Email Template or posts to CRM ``` This creates a powerful loop: ```text AI Voice Agent conversation ↓ Note created ↓ AI Assistant / Function / external endpoint responds ↓ Customer or internal team receives follow-up ``` --- ## Notes From Human Support Agents Human support agents can also use notes to intentionally trigger workflows. Examples: ```text "TRIGGER: retention_followup - customer wants discount before cancelling" ``` ```text "TRIGGER: supervisor_review - customer requested manager callback" ``` ```text "TRIGGER: shipping_resend_review - package not received" ``` A business can establish internal note conventions so Functions or AI Assistants know which notes should trigger which actions. This can make support workflows very flexible without requiring support agents to use complex forms. --- # Guidance for Notes and `note.created` Workflows When a user asks about notes, A RevCent workflow should consider both documentation and automation. The workflow should clarify: 1. Should this note be attached to the customer or to a related item such as sale, shipping, subscription, or renewal? 2. Is the note only for human history, or should it trigger action? 3. Should a `note.created` event trigger a Function? 4. Should a `note.created` event trigger an AI Assistant? 5. Should the note be sent to an external endpoint through a Function? 6. Should the note content be parsed for keywords, intent, sentiment, or next steps? 7. Should the note create or update metadata? 8. Should the note add the customer to a Customer Group? 9. Should the note trigger email or AI Voice Agent outreach? 10. Should a human review be required before action occurs? Important: ```text Notes can be passive history, but they can also be active workflow triggers. ``` --- ## Note Automation Guardrails Because notes can contain sensitive or ambiguous human language, automation should have guardrails. Recommended rules: - Do not trigger irreversible actions from ambiguous notes. - Use clear note conventions for workflow triggers. - Require human review for refunds, cancellations, legal issues, fraud/risk issues, chargebacks, and angry customers. - Avoid sending customer-facing messages from unclear note text. - Record what automation did in a follow-up note or metadata. - Do not expose private internal note text to the customer unless intentionally transformed into customer-safe language. - Do not allow notes to trigger repeated outreach without cooldown rules. - Use AI Assistant classification when note content needs interpretation. - Use Functions for deterministic actions and third-party endpoint calls. - Escalate uncertain cases to humans. --- ## Why Notes Matter Notes are one of the most flexible bridges between human judgment, AI reasoning, and automation. They let RevCent preserve context that is not always captured by structured fields. When combined with `note.created` event triggers, notes can also become the starting point for: - Customer follow-up - Support escalation - AI Assistant analysis - AI Voice Agent outreach - Email Template sending - CRM or helpdesk ticket creation - External endpoint notifications - Metadata updates - Customer group updates - Internal alerts - Customer success workflows This makes notes a critical part of the customer aspect of RevCent and a powerful way to turn support context into action. # Customer Portals Customer data can power Customer Portals. A Customer Portal can give customers access to self-service account experiences such as: - Viewing customer information - Managing payment methods - Reviewing orders or subscriptions - Handling account-related actions - Reducing support workload - Giving customers more control Because the portal is tied to RevCent customer data, it can provide a more useful and connected experience than a standalone form or generic support page. --- # Customer Engagement With RevCent Tools Customer data becomes more powerful when used across RevCent tools. ## Email Templates Customer data can be used to personalize emails. Examples: - First-name greeting - Purchase confirmation - Subscription renewal reminders - Payment failure notices - Refund confirmations - Shipping updates - Winback offers - VIP customer emails - Customer group-based messaging Customer metadata and purchase behavior can also support more personalized email content. --- ## AI Assistants AI Assistants can use customer-related data to help with: - Customer support - Customer analysis - Customer segmentation - Follow-up recommendations - Order and subscription context - Internal workflows - Filtering or qualification - Data lookup and action suggestions A customer stored in RevCent gives the AI Assistant more context than a standalone user message ever could. --- ## AI Voice Agents AI Voice Agents can use customer data to support phone workflows. Examples: - Inbound support with customer match - Outbound declined payment recovery - Subscription renewal recovery - Trial conversion calls - VIP customer follow-up - Shipping issue calls - Refund/cancellation triage - Post-purchase satisfaction calls - Winback/reorder calls A customer record enables more personalized and safer voice interactions. For example, an AI Voice Agent can: - Greet a verified customer by name. - Look up customer context. - Retrieve sale or subscription information when allowed. - Use customer groups to decide call behavior. - Use metadata for personalization. - Insert notes or metadata after the call. - Transfer or escalate based on customer value or status. --- ## Functions Functions can use customer data to connect RevCent with third-party systems. Examples: - Sync customer data to a CRM. - Pull customer data from a CRM before an AI workflow. - Enrich customer context from a data warehouse. - Validate loyalty status. - Generate custom offers. - Score churn risk. - Send customer events to external systems. - Create customer-specific logic for AI Assistants or AI Voice Agents. Because RevCent stores structured customer and purchase data, Functions can operate with meaningful business context. --- --- # Reporting With Google BigQuery and Customer Data RevCent customer data becomes especially powerful when analyzed through Google BigQuery. For metrics, counting, grouping, aggregation, and historical reporting, RevCent workflows should think in terms of BigQuery rather than list-style customer operations. BigQuery can combine customer data with related ecommerce records such as: - Sales - Product sales - Transactions - Shipments - Subscriptions - Subscription renewals - Trials - Refunds - Chargebacks - Fraud detections - Metadata - Customer groups - Notes - AI Voice Calls - AI Assistant activity, when available - Tracking/conversion data, when available This allows businesses to report not only on customers as individual records, but on customer behavior over time. --- ## Reporting on Customer Lifetime Values Customer lifetime values are a major reporting advantage. A customer can have lifetime value summaries for areas such as: - Total lifetime value - Initial sale value - Subscription renewal value - Gross amount - Net amount - Refunded amount - Discounted amount - Remaining amount - Amount to salvage - Number of sales - Average sale value - Number of subscription renewals - Overdue renewal counts - Chargeback totals - Fraud detection counts - Last sale date - Last subscription renewal date This makes it possible to answer questions like: ```text Which customers have the highest lifetime value? ``` ```text Which customer groups produce the most net revenue? ``` ```text Which campaigns create customers with the best lifetime value? ``` ```text Which customers have high sales but also high refunds? ``` ```text Which customers have overdue renewal value remaining? ``` ```text Which customers have chargebacks despite high lifetime value? ``` ```text Which customer metadata segments generate the most subscription renewal revenue? ``` --- ## Customer Reporting Examples BigQuery can help answer: | Business Question | Data Likely Needed | |---|---| | Which customers have the highest lifetime value? | Customer lifetime values. | | Which customer groups produce the most revenue? | Customer groups + sales/renewals/lifetime value. | | Which affiliates produce the best customers? | Customer metadata such as `affiliate_id` + lifetime value. | | Which customers are likely to reorder? | Product purchase dates + product purchase history. | | Which customers have not purchased recently? | Last sale date + lifetime value. | | Which subscription customers are at risk? | Subscription renewal history + overdue renewal counts. | | Which customers have high refund rates? | Refunded amount + sales totals. | | Which customers generate chargebacks? | Chargeback counts/amounts + customer history. | | Which products create repeat buyers? | Products purchased + customer purchase counts. | | Which customer segments should receive VIP support? | Lifetime value + groups + metadata. | --- --- # Secure Customer Payment Data and PCI-Compliant Storage A major advantage of storing customers in RevCent is that customer payment data can be securely stored and associated with the customer for future payment activity. RevCent is PCI compliant and stores customer payment data in a separate PCI storage layer. Raw credit card data is not exposed to RevCent users, customer support teams, AI Assistants, AI Voice Agents, or general customer workflows. Instead, RevCent exposes only generic, safe payment-method reference data that is useful for customer, billing, and payment workflows. --- ## Separate PCI Storage Layer Customer payment data is handled through RevCent’s secure PCI storage layer. This separation is important because it allows RevCent to support payment capabilities without exposing raw cardholder data throughout the platform. Conceptually: ```text Customer record ↓ References customer payment method ↓ Secure PCI storage layer holds sensitive payment data ↓ RevCent workflows use secure payment reference when payment processing is needed ``` The customer record can be associated with payment methods, but the sensitive raw card data remains protected in the PCI storage layer. --- ## Generic Payment Method Data Exposed RevCent can expose generic payment-method reference data for customer and billing context. Examples of generic customer card data include: - Customer card ID - Card type - First 6 digits / BIN - Last 4 digits - Expiration date - Default payment method status - Created/updated timestamps This generic information helps identify which stored payment method is associated with the customer without exposing raw credit card data. Example: ```text Default card ending in 4242 Visa expiring 08/28 Customer card ID: XXXXXXXXXXXXXXXXXXXX ``` This is useful for customer context, billing workflows, subscription management, and payment recovery while keeping sensitive payment data protected. --- ## Raw Credit Card Data Is Not Exposed Raw credit card data is not exposed to RevCent users, customer support teams, AI Assistants, AI Voice Agents, or general customer automation. The customer-facing and operational layers of RevCent may reference generic card details, but they do not expose the full underlying payment credentials. This means RevCent can support stored payment capabilities while keeping sensitive payment data isolated in the PCI-compliant payment layer. --- # How Secure Stored Payment Data Powers Ecommerce Because RevCent can securely associate payment methods with customers, businesses can support important ecommerce workflows that depend on stored customer payment data. --- ## Subscription Renewals Stored customer payment data allows RevCent to process subscription renewals without requiring the customer to re-enter payment information for every renewal. Conceptual flow: ```text Customer has subscription ↓ Customer has stored payment method ↓ Subscription renewal becomes due ↓ RevCent uses secure payment reference ↓ Renewal payment is attempted ``` This supports recurring revenue businesses by making renewal billing possible through securely stored customer payment methods. --- ## Trial Expiration Billing Stored customer payment data can also support trial expiration billing. Example: ```text Customer starts trial ↓ Payment method is securely stored ↓ Trial expires ↓ RevCent attempts payment using secure stored payment reference ``` This helps businesses convert trials into paid subscriptions or paid purchases without requiring a manual payment re-entry step at the moment of trial expiration. --- ## Decline Salvage and Payment Recovery Securely stored customer payment data also enables decline salvage and payment recovery workflows. Decline salvage means attempting to recover revenue from failed or declined payment attempts. Examples: - A subscription renewal declines. - A trial expiration payment declines. - A pending sale needs recovery. - A stored card may need to be retried according to business rules. - A customer may need to update their payment method. - A recovery workflow may send email or trigger AI Voice Agent outreach. Because the customer has a stored payment relationship, RevCent can support recovery workflows around that customer and payment method. Conceptual flow: ```text Payment attempt declines ↓ Customer and stored payment method remain associated ↓ RevCent can support retry, recovery, or update workflows ↓ Declined revenue may be salvaged ``` This is especially important for ecommerce businesses with subscriptions, trials, recurring billing, or payment recovery processes. --- ## Customer Lifetime Value and Stored Payment Data Stored payment methods also support more accurate customer value over time. When RevCent can securely process renewals, trials, and recovered payments, the customer record can accumulate richer lifetime value data. This can include: - Initial sale value - Subscription renewal value - Remaining amount - Amount to salvage - Refunded amount - Net value - Gross value - Renewal counts - Overdue renewal counts - Last sale date - Last subscription renewal date This makes stored customer payment data part of the broader customer intelligence layer. --- ## Why This Matters Secure customer payment storage gives RevCent customers a powerful foundation for ecommerce growth. It enables: - Recurring billing - Subscription renewals - Trial expiration billing - Payment retry strategies - Decline salvage - Pending sale recovery - Customer portal payment method management - Payment update workflows - Revenue recovery - More complete lifetime value tracking The key concept is: ```text RevCent securely stores customer payment data in a separate PCI storage layer, exposes only generic payment-method reference data, and uses the secure stored payment relationship to support renewals, trial billing, and decline salvage. ``` ## Metadata and Reporting Metadata is critical for customer reporting because it stores business-specific labels. Examples: ```text affiliate_id lead_source crm_id sales_rep customer_tier preferred_language campaign_category risk_score loyalty_status ``` A user may ask: ```text Show customer value by affiliate. ``` Technically, that may mean: ```text Group customer lifetime value by customer metadata where name = affiliate_id. ``` RevCent users and automation workflows should recognize that custom business terms often map to metadata and may need metadata discovery before building the correct BigQuery query. --- ## Customer Groups and Reporting Customer Groups make reporting easier because they create named customer segments. Examples: ```text Revenue by customer group Refund rate by customer group Chargebacks by customer group Average lifetime value by customer group Subscription renewal success by customer group Winback performance by customer group ``` Customer groups are useful when a segment should be reused across support, marketing, AI, and reporting. --- ## Reporting Can Drive Engagement BigQuery reporting should not be treated as a passive dashboard only. Reports can identify customers or segments that should trigger action. Examples: ```text BigQuery finds high-LTV customers with no purchase in 120 days ↓ AI Assistant reviews segment ↓ Email Template sends winback offer ``` ```text BigQuery finds customers with failed renewals and high remaining value ↓ AI Voice Agent calls customers ↓ Outcome saved to notes or metadata ``` ```text BigQuery finds customers who bought Product A three months ago ↓ Email Template sends reorder reminder ``` ```text BigQuery finds affiliate customers with high chargeback rates ↓ AI Assistant generates internal risk memo ``` The best customer workflows often combine: ```text BigQuery insight → AI decisioning → Email / Voice / Support action → customer record updated ``` --- ## BigQuery + AI Assistants AI Assistants can use BigQuery-style insights to drive customer strategy. Examples: - Summarize top customer segments by lifetime value. - Identify customers who should receive retention offers. - Recommend support prioritization based on LTV and risk. - Find patterns in refund-prone customer groups. - Suggest winback campaigns for inactive customers. - Analyze which metadata segments create the best customers. - Generate customer lists for targeted engagement. This gives businesses a powerful loop: ```text Customer data stored in RevCent ↓ BigQuery reports and aggregates behavior ↓ AI Assistant interprets or acts on the insight ↓ Email, voice, support, or Function workflow engages the customer ``` --- ## BigQuery Reporting Best Practice For customer reporting, RevCent workflows should prefer BigQuery when the user asks for: - Counts - Totals - Grouping - Filtering across many customers - Trends over time - Revenue by segment - Lifetime value by segment - Refund/chargeback analysis - Customer behavior analysis - Customer cohorts - Product purchase analysis - Subscription renewal analysis List/search operations are better for finding or inspecting specific customers. BigQuery is better for answering business questions across customer data. ## BigQuery Reporting Customer data is especially powerful in BigQuery. BigQuery can answer questions such as: - Which customer groups generate the most revenue? - Which metadata segments have the highest LTV? - Which customers bought a product more than once? - Which customers are at risk of churn? - Which customers had failed renewals? - Which customers have chargebacks? - Which campaigns generated high-value customers? - Which customers have high refunds relative to revenue? - Which products create repeat customers? - Which affiliates produce the best lifetime customers? The customer record is the anchor that makes these analyses valuable. --- ## Chargeback Mitigation Customer records can help support chargeback defense and investigation by connecting the customer to: - Originating sale - Tracking visitor details - Shipping details - Purchase history - Notes - Customer identity - Prior transactions - Refunds - Chargebacks - Fraud detections When customer data is centralized, it is easier to understand the full context around a disputed purchase. --- --- # Short-Term and Long-Term Customer Engagement Because RevCent stores customer identity, metadata, customer groups, and purchase behavior, it can support both short-term and long-term engagement strategies. Customer engagement should not be treated as a single email or one support call. A customer stored in RevCent can become part of an ongoing lifecycle: ```text Customer created ↓ Purchase behavior recorded ↓ Metadata and groups applied ↓ Email, AI, voice, support, and reporting workflows respond over time ↓ Customer profile becomes smarter with each interaction ``` --- ## Short-Term Engagement Short-term engagement is communication or automation that happens close to a customer event. Examples: - Customer created - Sale completed - Sale declined - Payment failed - Pending sale created - Shipment created - Shipment shipped - Shipment delivered - Subscription renewed - Subscription renewal failed - Trial is about to expire - Trial expiration failed - Refund created - Fraud detection created - Customer support interaction occurred Short-term engagement is useful because it responds while the customer action is fresh. Examples: ```text Customer purchases ↓ Send receipt or onboarding email ``` ```text Payment declines ↓ Send payment update email or trigger an AI Voice Agent recovery call ``` ```text Shipment is delivered ↓ Send post-purchase satisfaction email ``` ```text Subscription renewal fails ↓ AI Assistant reviews context or AI Voice Agent calls customer ``` Short-term engagement can help: - Confirm purchases. - Recover failed payments quickly. - Reduce support tickets. - Improve customer confidence. - Capture feedback. - Prevent chargebacks. - Guide customers through next steps. --- ## Long-Term Engagement Long-term engagement uses accumulated customer history over days, weeks, months, or years. Because RevCent stores purchase history, lifetime value, metadata, groups, subscriptions, renewals, notes, and product history, businesses can create engagement that becomes more intelligent over time. Examples: ```text Customer has not purchased in 90 days ↓ Add to winback segment ↓ Send personalized email or trigger AI Assistant workflow ``` ```text Customer has high lifetime value ↓ Add to VIP group ↓ Use elevated support and retention workflows ``` ```text Customer repeatedly buys a consumable product ↓ Trigger reorder reminder based on purchase dates ``` ```text Customer has failed subscription renewals ↓ Send long-term payment recovery sequence ``` ```text Customer bought Product A but not Product B ↓ Send targeted upsell or cross-sell campaign ``` Long-term engagement can support: - Winback campaigns - Reorder reminders - Subscription retention - Loyalty/VIP programs - Lifecycle marketing - Upsell and cross-sell campaigns - Renewal recovery - Lead nurture - Customer satisfaction programs - Churn prevention - Customer education - Loyalty tier upgrades - Internal account management --- # Email Templates for Customer Engagement Email Templates are one of the primary ways RevCent can engage customers using stored customer and event data. Email Templates can be used for both short-term transactional communication and long-term lifecycle engagement. --- ## Short-Term Email Template Use Cases Short-term Email Templates are usually event-driven. Examples: | Event | Email Template Use | |---|---| | Customer created | Welcome or account-created email. | | Sale success | Receipt, onboarding, fulfillment instructions. | | Sale decline | Payment failure or payment update instructions. | | Pending sale | Checkout reminder or payment completion email. | | Shipment shipped | Tracking notification. | | Shipment delivered | Delivery confirmation and feedback request. | | Subscription renewal success | Renewal confirmation. | | Subscription renewal decline | Payment update email. | | Trial upcoming / expired | Trial ending, conversion, or payment reminder. | | Refund created | Refund confirmation. | Because customer data is available, emails can be personalized with: - Name - Email - Products purchased - Order/sale context - Subscription context - Shipment status - Metadata - Customer group-based messaging - Custom arguments for API-direct sends --- ## Long-Term Email Template Use Cases Long-term Email Template usage is usually driven by segments, metadata, AI workflows, Functions, or reporting outputs. Examples: | Segment or Behavior | Email Engagement | |---|---| | No purchase in 90 days | Winback offer. | | Bought Product A | Cross-sell Product B. | | High lifetime value | VIP thank-you or exclusive offer. | | Subscription cancelled | Retention or feedback email. | | Trial did not convert | Education or special offer. | | Reorder window reached | Replenishment reminder. | | Affiliate customer group | Affiliate-specific messaging. | | Customer metadata indicates interest | Personalized content. | Long-term email engagement is powerful because RevCent can combine identity, purchase behavior, metadata, and customer groups to decide who should receive which message. --- ## API-Direct Email Engagement Email Templates with API-direct behavior can be used when another workflow explicitly sends the email. Examples: - AI Assistant decides a customer should receive a follow-up. - Function detects a customer qualifies for an offer. - Support agent triggers a personalized email. - AI Voice Agent sends an email after a call. - BigQuery/reporting workflow identifies a customer segment and triggers follow-up. API-direct templates are useful when the customer communication is not tied to one automatic event, but is instead triggered by a workflow decision. --- # AI Assistants for Customer Engagement AI Assistants can use stored customer data and business logic to create more intelligent engagement. An AI Assistant can run autonomously when triggered and can help decide what should happen next for customers or customer segments. --- ## Short-Term AI Assistant Use Cases Short-term AI Assistant workflows can respond to a recent event. Examples: ```text Sale declines ↓ AI Assistant reviews customer history, prior declines, metadata, and purchase value ↓ AI recommends or triggers the best recovery action ``` ```text Customer submits support request ↓ AI Assistant summarizes customer history and suggests next steps ``` ```text Fraud detection occurs ↓ AI Assistant reviews customer purchase history, metadata, and prior risk indicators ↓ AI prepares a human review summary ``` ```text Subscription renewal fails ↓ AI Assistant determines whether to send email, trigger AI Voice Agent, or escalate ``` Short-term AI Assistants can help with: - Support summaries - Payment recovery - PCI-compliant customer payment data storage in a separate secure payment layer, supporting subscription renewals, trial billing, and decline salvage decisions - Fraud/risk review - Customer prioritization - Post-sale next actions - Internal notifications - Data enrichment - Follow-up drafting --- ## Long-Term AI Assistant Use Cases Long-term AI Assistant workflows can analyze customer patterns over time. Examples: ```text Analyze customers with high LTV and no purchase in 120 days. ``` ```text Find customers likely to reorder based on purchase history. ``` ```text Identify subscription customers at risk of churn. ``` ```text Summarize customer segments by metadata, groups, and lifetime value. ``` ```text Recommend winback strategies for inactive VIP customers. ``` ```text Generate a list of customers who should receive a specific offer. ``` Long-term AI Assistants can help with: - Lifecycle marketing - Retention planning - Customer segmentation - Support prioritization - Revenue recovery strategy - Churn analysis - Offer generation - Customer summaries - Internal reporting memos - Next-best-action recommendations --- ## AI Assistants + Email Templates AI Assistants and Email Templates become especially powerful together. Example workflow: ```text AI Assistant identifies a customer segment ↓ AI Assistant determines recommended message or offer ↓ API-direct Email Template sends personalized follow-up ``` Examples: - AI Assistant finds high-LTV customers who have not purchased recently, then sends a winback email. - AI Assistant reviews failed renewal customers, then sends a payment update email. - AI Assistant identifies customers who bought Product A but not Product B, then sends a cross-sell email. - AI Assistant summarizes support history and sends a customer-specific resolution email. - AI Assistant identifies VIP customers and sends a personalized thank-you email. This creates a bridge between analysis and action. --- ## AI Assistants + AI Voice Agents AI Assistants can also work with AI Voice Agents. Example: ```text AI Assistant analyzes customer context ↓ AI Assistant determines whether a call is appropriate ↓ AI Voice Agent calls customer with specific instructions and context ↓ Call outcome is saved back to customer notes or metadata ``` Use cases: - Failed payment recovery - Subscription save calls - VIP follow-up - Trial conversion - Refund/cancellation triage - High-value customer support - Post-purchase satisfaction Stored customer data makes these AI-driven engagement workflows much more personalized and useful. # Customer Data Enables Personalization When RevCent stores customers and their behavior, businesses can personalize experiences based on real data. Examples: ```text A first-time customer gets a welcome email. A high-LTV customer gets VIP support. A subscription customer gets a renewal reminder. A customer with a failed payment gets an AI Voice Agent recovery call. A customer who bought Product A gets a targeted upsell for Product B. A customer with repeated refunds gets routed to manual review. A customer in a specific metadata segment gets a different offer. ``` This is where the “infinite possibilities” emerge. Once RevCent stores the customer and their purchase behavior, nearly every ecommerce workflow can become more intelligent, targeted, and automated. --- # Customer Data Enables Segmentation Customer segmentation can be based on: - Customer groups - Metadata - Campaign - Products purchased - Subscription status - Last purchase date - Lifetime value - Refund behavior - Chargeback behavior - Fraud detection behavior - Geography - Payment method/card data - Notes - AI-generated classifications - Third-party CRM data synced into metadata Examples: ```text Customers who bought Product A but not Product B Customers with LTV over $500 Customers who failed a renewal in the last 7 days Customers in VIP group with no purchase in 90 days Customers from affiliate_id = aff_123 with refund rate above threshold Customers with trial expired but no subscription renewal ``` These segments can power marketing, support, voice calls, retention workflows, and reporting. --- # Customer Data Enables Automation Once customer data exists in RevCent, automations become much more useful. Possible automations: - Add customers to groups after purchase. - Tag customers with metadata after behavior. - Send emails after a sale, refund, or renewal. - Trigger an AI Assistant for internal analysis. - Trigger an AI Voice Agent for payment recovery. - Run a Function to sync customer data to a CRM. - Create notes after support interactions. - Generate winback lists. - Identify high-risk customers. - Identify high-value customers. - Route subscriptions or renewals differently. - Create customer-specific follow-up workflows. Without a stored customer profile, these automations are much less powerful. With a stored customer profile, they can become highly targeted. --- --- # Customer Service and Support Teams The customer aspect of RevCent is especially valuable for customer service teams because it gives support agents and AI tools a centralized place to find, understand, and act on customer context. Great customer support depends on context. Without a connected customer record, a support representative may only see a single ticket, email, phone call, or order number. With RevCent, support teams can look at the customer as a whole: ```text Customer identity ↓ Purchase history ↓ Payment methods ↓ Subscriptions / trials ↓ Shipments ↓ Refunds ↓ Chargebacks ↓ Fraud detections ↓ Notes ↓ Metadata ↓ Customer groups ↓ AI / voice / email engagement history ``` This allows support to be faster, more accurate, and more personalized. --- ## Searching for Customers Customer support teams often start with a search. RevCent supports searching previously created customers using a search term. This allows a support agent, AI Assistant, or AI Voice Agent workflow to locate a customer when the support interaction begins with partial information. A customer may search by: - Name - Email - Phone - Internal customer ID - Metadata value - Address detail - Other searchable customer information Search results can include: - Customer ID - First name - Last name - Email - Phone - Address fields - Internal ID - Enabled status - Metadata - Matching highlights - Search score - Direct customer details URL This is important because customers rarely provide information in a perfect system format. A customer may say: ```text My name is Jane Smith. ``` or: ```text My email is jane@example.com. ``` or: ```text I ordered using my phone number. ``` or: ```text My order was under my company name. ``` or: ```text I have a CRM/customer ID from another system. ``` Search lets the support team find the correct RevCent customer record and then use the customer context for better support. --- ## Human Customer Support Workflow A human customer support representative can use RevCent customer data to quickly understand the customer’s full relationship with the business. Example workflow: ```text Customer contacts support ↓ Support searches for customer by email, phone, name, or internal ID ↓ Support opens customer record ↓ Support reviews status, groups, metadata, notes, cards, purchases, subscriptions, shipments, refunds, chargebacks, and lifetime value ↓ Support resolves issue or escalates with complete context ↓ Support adds a note or updates customer metadata/group if appropriate ``` This helps support teams avoid asking the customer to repeat information the business already knows. --- ## Support Context Available on a Customer A support team may use customer context such as: | Context | How It Helps Support | |---|---| | Contact information | Confirms identity and communication channel. | | Customer status | Shows whether the customer is enabled, disabled, or blocked. | | Customer groups | Identifies VIPs, wholesale customers, at-risk customers, support priority, etc. | | Metadata | Shows custom business context such as affiliate, CRM ID, sales rep, customer tier, or risk flag. | | Payment methods | Helps answer billing/payment questions using safe card reference data such as card type, first 6, last 4, and expiration. | | Products purchased | Helps support answer product, reorder, warranty, or upsell questions. | | Sales and transactions | Helps support investigate charges, payments, failed payments, and refunds. | | Shipments | Helps support answer tracking, delivery, and fulfillment questions. | | Subscriptions and renewals | Helps support answer recurring billing, cancellation, renewal, and payment failure questions. | | Trials | Helps support explain trial status or trial expiration billing. | | Notes | Gives human and AI agents prior interaction history. | | Lifetime value | Helps prioritize high-value customers or tailor escalation rules. | | Chargebacks and fraud detections | Helps support handle disputes, risk, and sensitive interactions carefully. | --- ## Superior Support Through Connected Purchase Behavior Because RevCent stores customer purchase behavior, a support agent can answer questions with much more confidence. Examples: ```text “I see you purchased Product A twice, most recently on March 12.” ``` ```text “Your subscription renewal failed, but your subscription is still associated with this account.” ``` ```text “I see this shipment is tied to your most recent sale.” ``` ```text “I see you were already refunded for that product sale.” ``` ```text “I see a note from the previous call, so I can pick up where that conversation left off.” ``` The support experience becomes more personal and less repetitive. --- ## Using Customer Groups for Support Prioritization Customer Groups can be used by support teams to quickly understand how a customer should be handled. Examples: ```text VIP Wholesale Subscription Save Fraud Review Chargeback Risk Manual Review High LTV Trial Conversion Escalated Support ``` Support behavior can differ by group. Examples: - VIP customers may be escalated faster. - Wholesale customers may require different policies. - Subscription Save customers may be routed to a retention workflow. - Fraud Review customers may require manual verification. - High-LTV customers may receive higher-touch service. - Trial Conversion customers may receive a different follow-up offer. Customer groups make support workflows easier to standardize. --- ## Using Metadata for Custom Support Context Metadata can store the support context that is unique to each business. Examples: ```text support_priority = high assigned_agent = alex crm_id = 12345 loyalty_tier = gold preferred_language = es last_support_outcome = replacement_sent refund_policy_exception = approved risk_review_status = pending ``` This allows support teams and AI tools to adapt to business-specific workflows without requiring every custom field to become a permanent built-in customer field. --- ## Notes as Support Memory Customer notes are valuable because they preserve support memory. Notes can capture: - What the customer asked about. - What support told the customer. - Whether a refund was promised. - Whether a replacement was sent. - Whether a customer requested cancellation. - Whether the customer was upset. - Whether the case was escalated. - Whether a human reviewed a fraud/risk issue. - Whether an AI Voice Agent called the customer. - Whether payment recovery was attempted. - Whether the customer requested a callback. This creates continuity across support channels. A different agent, AI Assistant, or AI Voice Agent can understand what happened previously. --- # AI Voice Agents for Customer Support AI Voice Agents can use RevCent customer data to provide customer support by phone. This is useful for both inbound and outbound support workflows. --- ## Inbound AI Voice Support For inbound calls, an AI Voice Agent can help customers who call into a support number. If the caller can be matched to a customer, the AI Voice Agent can use customer context to provide more useful support. Inbound AI Voice Agent examples: - “Where is my order?” - “I need help with my subscription.” - “Why was I charged?” - “Can I update my payment method?” - “Can I cancel my subscription?” - “Can I speak to a human?” - “Did my refund go through?” - “When will my package arrive?” When allowed by the agent’s enabled system actions, an AI Voice Agent can retrieve customer, sale, shipment, transaction, subscription, renewal, or trial details to answer questions. Example flow: ```text Customer calls support number ↓ AI Voice Agent attempts customer match by phone ↓ If matched, agent uses customer context carefully ↓ Agent verifies identity when needed ↓ Agent answers routine questions or retrieves allowed data ↓ Agent creates note or metadata outcome ↓ Agent transfers to a human when required ``` --- ## Outbound AI Voice Support AI Voice Agents can also support outbound workflows. Examples: - Failed payment recovery - Pending sale recovery - Subscription renewal recovery - Trial conversion follow-up - Shipping issue follow-up - VIP customer follow-up - Post-purchase satisfaction calls - Winback or reorder calls - Refund/cancellation triage - Internal escalation calls for high-value customer issues Because the customer record connects purchase behavior, payment context, and contact information, outbound calls can be more targeted and more useful. Example: ```text Subscription renewal fails ↓ Customer record provides phone, subscription, renewal, customer group, and metadata context ↓ AI Voice Agent calls customer ↓ Agent explains the issue and helps update payment if allowed ↓ Agent creates note and metadata with call outcome ``` --- ## AI Voice Agents Should Use Customer Context Carefully AI Voice Agents should not simply reveal all customer information. They should follow rules for: - Customer verification - Privacy - Payment handling - Refund behavior - Subscription changes - Escalation - Transfer to human - End-call behavior - Notes and metadata - Allowed system actions Customer data makes AI Voice Agents more powerful, but instructions and enabled actions determine what the agent can safely do. For example: - Include `SearchCustomers` if the AI Voice Agent should search for an unmatched inbound caller. - Include `GetCustomer` if the AI Voice Agent should retrieve current customer details. - Include `GetSale` if the AI Voice Agent should answer order/sale questions. - Include `GetShipment` if the AI Voice Agent should answer shipping questions. - Include `GetSubscription` or `GetSubscriptionRenewal` for subscription workflows. - Include `CreateNote` if the AI Voice Agent should record call outcomes. The AI Voice Agent should only use system actions that are needed for the support workflow. --- # Human Support and AI Voice Agents Working Together RevCent customer data allows human support and AI Voice Agents to work together instead of operating in separate silos. Examples: ## AI Handles Routine Questions, Human Handles Exceptions ```text AI Voice Agent answers order status, subscription status, and payment failure questions ↓ Complex or upset customer requests human ↓ AI transfers call ↓ Human sees customer context and notes from the AI call ``` ## Human Support Creates Context, AI Uses It Later ```text Human support adds note: "Customer requested callback about failed renewal" ↓ AI Voice Agent later calls customer with appropriate context ↓ AI creates note with outcome ``` ## AI Creates Structured Follow-Up for Humans ```text AI Voice Agent handles inbound support call ↓ Customer requests exception or policy review ↓ AI creates note and metadata ↓ Human support reviews with full customer history ``` ## Support Uses Customer Groups to Control AI Behavior ```text Customer is in VIP group ↓ AI Voice Agent uses VIP escalation rules ↓ Human support receives high-priority transfer or note ``` --- # Customer Search as the Start of Better Support Customer search is often the first step in a support workflow. Support may begin with: ```text Search by email Search by phone Search by name Search by internal customer ID Search by metadata ``` Then the support or AI workflow can move from search result to full customer context. This is especially useful when the customer: - Uses a different email than expected. - Calls from a different phone number. - Provides only partial identifying information. - Has multiple orders. - Has an internal CRM ID. - Is associated with metadata from a lead source, affiliate, or campaign. - Needs support across sales, subscriptions, and shipments. Search reduces friction and helps support teams find the right customer faster. --- # Superior Customer Support Outcomes When customer service teams use RevCent as the customer source of truth, they can provide better support outcomes. Benefits include: - Faster customer lookup. - Less repetitive questioning. - More personalized service. - Better order and subscription context. - Better payment issue resolution. - Better refund and cancellation handling. - Better escalation decisions. - Better VIP handling. - Better fraud/risk awareness. - Better continuity between human and AI support. - Better support notes and audit history. - Better follow-up through email, AI Assistants, AI Voice Agents, or Functions. The result is a support system where every interaction can become more informed because it is connected to the customer’s actual ecommerce history. # Customer Data Enables Better Support Support teams need context. A customer stored in RevCent gives support teams a connected view of: - Who the customer is - What they bought - When they bought it - What they paid - Whether they were refunded - Whether they have subscriptions - Whether they have failed renewals - Whether they have open shipment issues - Whether they have chargebacks - Whether they have fraud detections - Whether they have notes - Whether they are VIP, blocked, or part of a special group This can reduce repeated questions and improve support quality. It can also help AI tools give better assistance. --- # Customer Data Enables Revenue Recovery Customer data is essential for revenue recovery. Examples: - Failed payment recovery - Pending sale recovery - Subscription renewal recovery - Trial conversion - Salvage transaction workflows - Reorder reminders - Winback campaigns - Abandoned checkout follow-up - AI Voice Agent recovery calls - Targeted email sequences Because RevCent can connect customer identity to purchase/payment behavior, revenue recovery can be based on actual customer context rather than generic outreach. --- # Customer Data Enables Smarter AI AI is only as useful as the context it can access. When customers and purchase behavior are stored in RevCent, AI tools can reason with meaningful ecommerce context. AI can help answer: - What is this customer’s history? - Is this a VIP customer? - What products has this customer purchased? - Has this customer refunded before? - Does this customer have a failed renewal? - Should this customer receive a retention offer? - Should this customer be escalated to a human? - Is this customer part of a special group? - What metadata exists for this customer? - What action should happen next? This enables AI Assistants, AI Voice Agents, and Functions to become business-aware instead of generic. --- --- # Independent AI Agents Connected Through RevCent Independent AI Agents can connect to RevCent through integrations and use customer data as an ongoing source of business context. This means an AI Agent does not need to only respond when a human asks a question. It can be configured to repeatedly check customer-related data, detect important conditions, and recommend or conduct outreach when appropriate. Conceptually: ```text Independent AI Agent ↓ Connects to RevCent ↓ Reads customer, purchase, subscription, payment, metadata, group, and reporting data ↓ Detects conditions that matter to the business ↓ Decides whether outreach or internal action is needed ↓ Uses Email Templates, AI Assistants, AI Voice Agents, Functions, notes, metadata, or customer groups ↓ Stores the result back in RevCent ``` This turns RevCent customer data into an active engagement system instead of a passive database. --- ## What Independent AI Agents Can Monitor Independent AI Agents can monitor many customer-related signals. Examples: - Newly created customers. - Customers with recent purchases. - Customers with declined sales. - Customers with pending sales. - Customers with failed subscription renewals. - Customers with trials nearing expiration. - Customers with trials that expired but did not convert. - Customers with high lifetime value. - Customers with no recent purchase. - Customers with repeated refunds. - Customers with chargebacks. - Customers with fraud detections. - Customers added to specific customer groups. - Customers with specific metadata values. - Customers with missing metadata. - Customers with support notes requiring follow-up. - Customers who purchased a specific product. - Customers who have not reordered after an expected interval. - Customers whose lifetime value or behavior changed significantly. Because RevCent stores both customer identity and customer behavior, an independent AI Agent can monitor not just “who the customer is,” but also “what the customer is doing over time.” --- ## Short-Term Monitoring Short-term monitoring focuses on recent events that may require quick action. Examples: ```text Sale declined ↓ AI Agent checks customer history and value ↓ AI Agent determines whether to send an email, trigger an AI Voice Agent, or add the customer to a recovery group ``` ```text Customer created ↓ AI Agent checks source metadata and campaign ↓ AI Agent sends a welcome email or assigns a customer group ``` ```text Subscription renewal failed ↓ AI Agent checks customer value and prior renewal history ↓ AI Agent triggers recovery outreach ``` ```text Shipment delivered ↓ AI Agent waits appropriate time ↓ AI Agent sends satisfaction or reorder follow-up ``` Short-term monitoring is useful for fast reaction. It can help recover revenue, reduce support issues, and provide timely customer communication. --- ## Long-Term Monitoring Long-term monitoring looks for patterns over days, weeks, months, or years. Examples: ```text Customer has not purchased in 90 days ↓ AI Agent evaluates lifetime value and prior products ↓ AI Agent sends winback offer or adds customer to winback group ``` ```text Customer buys consumable product every 45 days ↓ AI Agent detects reorder window ↓ AI Agent sends reorder reminder ``` ```text Customer has high LTV but recent refund issue ↓ AI Agent creates internal note or escalates to support ``` ```text Customer group has rising refund rate ↓ AI Agent generates internal report ↓ AI Agent recommends support or product changes ``` Long-term monitoring is useful for retention, lifecycle marketing, customer success, churn prevention, and strategic reporting. --- # AI Agent Outreach Capabilities When an AI Agent determines that outreach is needed, it can use several RevCent tools. ## Email Outreach AI Agents can use Email Templates for customer communication. Examples: - Welcome emails - Payment recovery - PCI-compliant customer payment data storage in a separate secure payment layer, supporting subscription renewals, trial billing, and decline salvage emails - Subscription renewal reminders - Trial conversion emails - Reorder reminders - Winback offers - VIP thank-you emails - Post-purchase education - Support follow-up emails Email Templates are useful when the outreach should be written, branded, and repeatable. --- ## AI Voice Agent Outreach AI Agents can trigger or coordinate AI Voice Agent workflows when a phone call is more appropriate than an email. Examples: - High-value declined payment recovery - Subscription save calls - VIP customer follow-up - Trial conversion calls - Refund/cancellation triage - Shipping issue follow-up - Post-purchase satisfaction calls - Winback calls AI Voice Agents are especially useful when: - The issue is urgent. - The customer value is high. - A conversation is more effective than email. - The customer may need help completing payment. - The business wants a more personal experience. --- ## AI Assistant Workflows Independent AI Agents can use AI Assistants for analysis, decisioning, and internal tasks. Examples: - Summarize customer history before outreach. - Generate next-best-action recommendations. - Identify which customers should receive an offer. - Analyze failed renewal customers. - Create internal customer success summaries. - Decide whether email or phone outreach is more appropriate. - Draft personalized messaging for review. - Produce daily or weekly customer engagement reports. AI Assistants are useful when the AI Agent needs reasoning, summarization, or decision support before taking action. --- ## Functions and Third-Party Integrations AI Agents can use Functions when outreach or monitoring requires custom logic or third-party systems. Examples: - Sync customer status to a CRM. - Pull loyalty status from another platform. - Send customer data to a helpdesk. - Create a ticket in an external support system. - Check external shipping or fulfillment status. - Generate a custom offer. - Call a third-party marketing platform. - Enrich customer metadata from another source. This allows independent AI Agents to use RevCent customer data as the center of a broader ecommerce automation ecosystem. --- ## Notes, Metadata, and Customer Groups as Feedback Loops Outreach should usually leave a record. AI Agents can use customer notes, metadata, and customer groups to store outcomes. Examples: ```text ai_agent_last_reviewed_at = 2026-05-28 ai_agent_recommended_action = payment_recovery_call ai_agent_outreach_status = email_sent ai_agent_outreach_status = voice_call_completed ai_agent_outreach_outcome = customer_paid ai_agent_outreach_outcome = no_answer ai_agent_outreach_outcome = escalated_to_human ``` Customer groups can also represent workflow states: ```text Payment Recovery Winback Candidate VIP Follow-Up Needs Human Review Subscription Save Reorder Reminder ``` This creates a closed loop: ```text Monitor → Decide → Outreach → Record outcome → Use outcome for future decisions ``` --- # Independent AI Agent Examples ## Revenue Recovery Agent Purpose: ```text Continually monitor failed payments, declined sales, pending sales, and overdue renewals. ``` Possible actions: - Send payment recovery email. - Trigger AI Voice Agent call for high-value customers. - Add customer to Payment Recovery group. - Insert metadata with recovery status. - Create note for human follow-up. --- ## Retention Agent Purpose: ```text Monitor customer behavior for churn risk and retention opportunities. ``` Possible signals: - Failed renewal - Subscription cancellation - No purchase in expected timeframe - High LTV customer becoming inactive - Trial expired without conversion Possible actions: - Send retention email. - Trigger AI Assistant to generate an offer. - Trigger AI Voice Agent for high-value customers. - Add customer to Subscription Save or Winback group. --- ## Reorder Agent Purpose: ```text Monitor purchase history and identify likely reorder windows. ``` Possible signals: - Product purchase dates - Product type - Average reorder interval - Customer metadata - Last sale date Possible actions: - Send reorder reminder. - Add customer to Reorder Candidate group. - Create personalized offer. - Trigger AI Assistant to select recommended product. --- ## VIP Customer Success Agent Purpose: ```text Monitor high-value customers and make sure important issues receive priority. ``` Possible signals: - High lifetime value - VIP customer group - Recent refund - Failed payment - Support note - Chargeback risk - Delayed shipment Possible actions: - Create internal support note. - Trigger human escalation. - Send VIP follow-up email. - Trigger AI Voice Agent for personal outreach. - Add metadata indicating customer success review. --- ## Risk and Support Escalation Agent Purpose: ```text Monitor customers for support, fraud, refund, or chargeback risk. ``` Possible signals: - Repeated refunds - Chargeback created - Fraud detection - Support notes with escalation language - Blocked or disabled status - High refund-to-revenue ratio Possible actions: - Add customer to Fraud Review or Manual Review group. - Create internal summary. - Notify support team. - Prevent automated outreach if inappropriate. - Trigger AI Assistant review. --- # Guardrails for Independent AI Agents Independent AI Agents can be powerful, but they should operate with clear guardrails. RevCent workflows should ensure AI Agents have: - Clear purpose. - Clear monitoring criteria. - Clear outreach rules. - Clear limits on frequency. - Clear stop conditions. - Clear escalation rules. - Clear privacy and verification rules. - Clear rules for when not to contact a customer. - Clear rules for sensitive payment-related situations. - Clear recording of outcomes in notes, metadata, or groups. Examples of important guardrails: ```text Do not call customers outside allowed hours. Do not contact customers who opted out. Do not repeatedly contact the same customer without a cooldown. Do not expose sensitive customer or payment data. Do not make refund, cancellation, or payment changes unless explicitly allowed. Escalate angry, confused, legal, fraud, or chargeback-related situations to a human. Record every outreach attempt. ``` --- # Guidance for Independent AI Agents When a user wants an independent AI Agent to monitor customer data, RevCent workflows should clarify: 1. What customer condition should be monitored? 2. Should the monitoring be short-term, long-term, or both? 3. How often should the agent check? 4. Which customer data should be considered? 5. Should BigQuery be used for segment/metric detection? 6. What action should happen when the condition is met? 7. Should outreach be email, voice, AI Assistant, Function, human task, or customer group update? 8. What limits should prevent over-contacting? 9. What customer groups or metadata should suppress outreach? 10. What notes, metadata, or groups should record the outcome? 11. When should a human be notified or take over? 12. Should the workflow be reviewed before being enabled? Important: ```text Independent AI Agents should not simply “contact customers.” They should monitor specific customer conditions, take specific approved actions, and record outcomes. ``` --- # Why This Matters Independent AI Agents connected through RevCent integrations can turn RevCent customer data into an always-on customer engagement layer. Instead of waiting for a human to manually run reports, search for customers, or trigger outreach, AI Agents can continuously watch for meaningful customer moments. This creates a system where RevCent customer data can drive: - Revenue recovery - Customer retention - Support escalation - Reorder campaigns - Winback campaigns - VIP customer success - Fraud/risk review - Lifecycle marketing - Personalized outreach - Internal customer intelligence The value comes from the combination of: ```text Stored customer data + purchase behavior + metadata + customer groups + lifetime values + BigQuery reporting + AI Assistants + AI Voice Agents + Email Templates + Functions + RevCent-connected independent agents ``` Together, these capabilities allow ecommerce businesses to build customer engagement systems that are proactive instead of reactive. # Infinite Possibilities From Connected Customer Data The value of the customer aspect of RevCent compounds because customer data connects to many other RevCent systems. When RevCent stores the customer and their purchase behavior, a business can create nearly unlimited workflows around that customer. Examples: - Personalized emails based on products purchased. - AI Voice Agent calls based on failed payments. - Customer portal experiences based on customer status. - Payment routing based on customer group. - Subscription save offers based on lifetime value. - Fraud review based on chargeback history. - BigQuery dashboards by customer segment. - CRM sync based on metadata. - VIP escalation based on customer group. - Reorder campaigns based on purchase dates. - Refund policy workflows based on purchase/refund history. - AI-generated customer summaries for support agents. - Internal alerts for high-value customer issues. - Customer-specific offers generated by Functions. - Retention campaigns based on subscription behavior. This is why storing customers in RevCent is so powerful: the customer becomes the common object that connects commerce, payments, support, AI, automation, reporting, and engagement. --- # Guidance for Customer-Related Workflows When users ask about customers, RevCent workflows should think beyond basic customer creation or lookup. A RevCent workflow should consider: 1. Is the user asking about an individual customer, a segment, or a behavior pattern? 2. Is the relevant data stored in standard customer fields, customer groups, metadata, purchase history, or related records? 3. Is the user really asking about metadata without saying “metadata”? 4. Should BigQuery be used for metrics, counting, grouping, or historical analysis? 5. Should customer groups be used for reusable segmentation? 6. Should metadata be used for custom business labels? 7. Should AI Assistants, AI Voice Agents, Email Templates, or Functions be used for engagement? 8. Should a Customer Portal be part of the customer experience? 9. Should sensitive payment data be handled only through secure customer card workflows? 10. Could customer status, blocking, or groups affect what should happen next? --- # Best Practices ## Use Customer Groups for Clear Segments Use customer groups when a segment should be reusable and easy to understand. Examples: ```text VIP Wholesale Subscription Save Fraud Review High LTV Trial Conversion Winback ``` --- ## Use Metadata for Custom Business Context Use metadata for flexible custom data. Examples: ```text affiliate_id crm_id sales_rep lead_source customer_tier preferred_language last_offer_sent risk_score ``` --- ## Keep Customer Data Accurate Accurate customer data improves: - Support - Billing - Shipping - Reporting - AI workflows - Customer portals - Revenue recovery - Fraud prevention --- ## Use BigQuery for Metrics For counting, grouping, aggregations, and reporting, use BigQuery rather than list operations. Examples: ```text Count customers by group Revenue by customer metadata Average LTV by campaign Refund rate by customer segment ``` --- ## Avoid Treating Customer Data as Flat A customer is not just a single object. A customer is connected to many related objects: ```text Customer → Sales → Product Sales → Shipments → Transactions → Subscriptions → Renewals → Refunds → Chargebacks → Metadata → Notes ``` Good RevCent workflow behavior means understanding these relationships. --- # Example Customer-Centered Workflows ## VIP Support ```text Customer has high lifetime value ↓ Add to VIP group ↓ AI Voice Agent or support workflow treats customer as priority ↓ Create note after interaction ``` ## Failed Payment Recovery ```text Subscription renewal fails ↓ Customer record shows contact info and payment context ↓ AI Voice Agent calls customer ↓ Customer updates payment method ↓ Renewal is recovered ``` ## Personalized Email ```text Customer purchased Product A ↓ Metadata or purchase history identifies upsell opportunity ↓ Email Template sends personalized Product B offer ``` ## Reorder Campaign ```text Customer bought consumable product ↓ Purchase date indicates reorder window ↓ Customer added to reorder group ↓ Email, AI Assistant, or AI Voice Agent follows up ``` ## Chargeback Context ```text Chargeback occurs ↓ Customer record links to sale, product, tracking, notes, and prior behavior ↓ Business reviews full context ↓ Chargeback mitigation document can be generated when appropriate ``` --- # Summary The customer aspect of RevCent is foundational because it gives RevCent a durable customer object that connects identity, payment methods, purchase behavior, metadata, groups, notes, and related ecommerce records. This unlocks superior capabilities across: - Segmentation through highly granular Customer Groups and qualifier-based customer lifecycle states - Personalization - Customer support - Payment recovery - PCI-compliant customer payment data storage in a separate secure payment layer, supporting subscription renewals, trial billing, and decline salvage - Subscription management - Customer portals - AI Assistants - AI Voice Agents - Functions - Email Templates - BigQuery reporting using customer data, customer metadata, groups, purchase behavior, and lifetime values - Short-term and long-term engagement through Email Templates, AI Assistants, AI Voice Agents, and Functions - Chargeback mitigation - Fraud/risk workflows - Lifecycle automation triggered by customer group added/removed events - Independent AI Agents connected through RevCent integrations for continuous monitoring and outreach - Notes and `note.created` event workflows for Functions, AI Assistants, AI Voice Agents, Email Templates, metadata, customer groups, and external endpoint actions The most important RevCent workflow concept is: ```text Once RevCent stores a customer and their purchase behavior, that customer becomes the anchor for nearly unlimited ecommerce workflows. ``` --- Document Parent Directory * [Operations](https://revcent.com/documentation/markdown/mcp/operation/index.md) - AI/MCP details and overviews for operations available within the RevCent MCP.