“Most platforms automate marketing. Fundle automates outcomes — incremental revenue, retention lift, attributed footfall. The number is the product.”
- •Understand why fragmented multi-brand loyalty programs bleed retention and revenue in Indian malls
- •Explore how agentic AI autonomously stitches customer journeys across 270+ partner brands
- •Compare legacy loyalty stacks against AI-native platforms built for India's retail complexity
- •Follow a five-step playbook to deploy AI loyalty agents across your mall or retail chain
- •Measure the KPIs that actually matter: redemption rate, cross-brand visit frequency, and CLV uplift
India's organised retail sector is at an inflection point. Mall footfall crossed 800 million visits annually across the top 50 Grade-A malls in FY2024, yet average loyalty programme enrolment sits below 18% of unique visitors — a gap that costs mall operators and their brand tenants somewhere between ₹2,000 and ₹4,500 crore in unrealised repeat-purchase revenue every year. The mechanics of why this happens are not mysterious: when a customer shops at Tanishq on the ground floor, picks up a pair of frames from Lenskart on the first floor, and grabs a coffee from Cafe Coffee Day before leaving, three separate loyalty ledgers record three separate transactions. Nobody connects the dots. Nobody sends that customer a message that says, 'You just earned 340 points across three stores — redeem ₹170 off your next visit to FabIndia.' That moment of unified delight never happens.
The problem is structural. Legacy loyalty platforms — whether homegrown Excel-and-SMS stacks or first-generation SaaS tools like EasyRewardz or Capillary's older modules — were architected around single-brand or single-tenant logic. They can issue points, they can send bulk SMS blasts, and they can produce monthly cohort reports. What they cannot do is reason across brands in real time, personalise offers at the moment of checkout, or autonomously escalate a churning customer to a retention intervention without a human analyst pulling a query at 11 PM.
This is exactly the problem that AI loyalty agents for customer engagement are designed to solve. Unlike traditional rules-based engines, an AI loyalty agent operates with goal-directed autonomy: it perceives context (transaction data, location signals, browsing history, weather, occasion calendars), reasons about the best next action, executes that action across channels — push notification, WhatsApp, in-app offer, POS prompt — and learns from the outcome to refine its next move. It is not a chatbot. It is not a recommendation widget. It is a persistent, always-on agent that manages the customer relationship on behalf of the brand or mall operator.
Fundle was built from the ground up to operationalise this vision inside India's multi-brand retail environment. Where competitors offer point-and-shoot campaign tools, Fundle AI Platform fields a network of specialised AI agents — each responsible for a slice of the customer lifecycle — that coordinate with each other the way a well-run mall operations team coordinates across tenants. The sections that follow unpack why the timing is right, what good looks like in practice, and how operators can move from fragmented loyalty chaos to compound customer value.
The Multi-Brand Loyalty Gap in Indian Retail — By the Numbers
Challenges of Multi-Brand Loyalty Programs in Indian Malls
Running a loyalty programme inside a mall is not the same as running one for a single brand. A mall CMO at Phoenix Marketcity Mumbai or Select CITYWALK Delhi is simultaneously the convener, the infrastructure provider, and the neutral broker among 150 to 300 brand tenants — each of whom has their own POS system (POSist, Petpooja, GoFrugal, Wondersoft, or a proprietary stack), their own CRM preferences, and their own promotional calendar. Getting Lifestyle, Pantaloons, Manyavar, and Apollo Pharmacy to agree on a unified points currency is a governance problem before it is a technology problem.
The technology layer compounds the challenge. POS integrations across heterogeneous systems are notoriously fragile. A tenant on GoFrugal may send transaction webhooks in a different schema than a tenant on Wondersoft. Batch settlement files arrive at different times of day. SKU-level data is often missing or inconsistent, which makes category-level personalisation nearly impossible. The result: most mall loyalty programmes end up operating as thin points-accrual wrappers over what is essentially a fragmented transaction database. Members earn points they never redeem — India's average loyalty point redemption rate in retail sits around 28%, compared to 60%+ in mature markets like South Korea or the UAE.
Data privacy and consent management add another layer of friction. Post-DPDP Act 2023, Indian operators must obtain explicit, purpose-specific consent before using customer data for cross-brand targeting. A customer who consented to Reliance Trends' communications did not automatically consent to receiving offers from every other tenant in the mall. Stitching consent records across brands, audit-trail compliant, is a capability most legacy platforms do not have.
Finally, there is the measurement problem. Mall marketing teams typically report on footfall, dwell time, and tenant sales — metrics sourced from three different vendors who rarely reconcile with each other. Loyalty ROI is reported in vanity metrics: total points issued, total members enrolled. The questions that actually matter — which cross-brand journeys drive the highest basket size, which segment is one bad experience away from churning, which offer type drives a second visit within 14 days — go unanswered because the analytical plumbing does not exist. This is the gap that agentic AI in retail loyalty was purpose-built to fill.
The Multi-Brand Customer Loyalty Funnel — Where Indian Malls Lose Value
How Agentic AI Facilitates Unified Engagement Across Brands
The term 'agentic AI' has entered retail vocabulary quickly, but it is worth being precise about what it means in a loyalty context. An AI agent is a system that can perceive its environment, set sub-goals toward a larger objective, execute actions through APIs and integrations, evaluate results, and update its strategy — all without a human triggering each step. In a multi-brand retail setting, this means the agent can autonomously detect that a customer who normally visits every 12 days has not appeared in 18 days, calculate the optimal win-back offer based on that customer's historical category preferences and price sensitivity, dispatch a personalised WhatsApp message with a time-bound cross-brand voucher, monitor whether the customer clicks and visits, and update the churn-risk score accordingly. No analyst, no campaign brief, no approval workflow.
This is fundamentally different from what MoEngage, WebEngage, or Xeno do. Those are excellent channel-execution platforms — they can orchestrate journeys and segment audiences — but they depend on a human marketer to define the segments, write the journeys, and set the rules. The agentic layer sits above that: it decides what the journey should be, then instructs the execution platform. Think of it as a Chief Loyalty Officer that works 24 hours a day across every customer simultaneously.
For a mall operator, agentic AI in retail loyalty enables three capabilities that were previously impossible at scale. First, real-time cross-brand stitching: the moment a customer's transaction clears at Manyavar, the agent knows that the same customer bought ethnic accessories at FabIndia six days ago and is 80% likely to be preparing for a wedding. It can surface a personalised offer from a relevant beauty or gifting tenant before the customer leaves the mall floor. Second, autonomous tier management: rather than waiting for a quarterly batch job to upgrade a customer to Gold tier, the agent does it the moment the threshold is crossed and immediately communicates the new benefits. Third, consent-aware personalisation: the agent maintains a real-time consent graph per customer, ensuring that cross-brand data usage stays within the permissions granted, producing an audit log compliant with India's DPDP Act.
The compounding effect is significant. Customers who experience seamless cross-brand recognition — where the mall 'knows' them regardless of which store they enter — show a 2.8× higher 90-day retention rate in Indian retail pilots compared to customers in standard single-brand programmes. The data advantage grows over time: every agent decision generates a labelled training signal that improves future recommendations, creating a flywheel that a rules-based system can never replicate.
Legacy Loyalty Platforms vs. AI Loyalty Agents: What Indian Mall CMOs Actually Get
Fundle's Multi-Brand AI Loyalty Framework for Indian Retail
Fundle connects 270+ partner brands across malls through AI-powered loyalty engagement — a network effect that no single-brand programme can replicate. The Fundle AI Platform is structured around three interlocking layers that together constitute the most complete agentic loyalty architecture available for Indian multi-brand retail today.
The first layer is the Fundle Loyalty data fabric. This is the integration and identity resolution layer that ingests transaction signals from heterogeneous POS systems — POSist, GoFrugal, Wondersoft, Petpooja, proprietary stacks — normalises them into a canonical event schema, and resolves them against a unified customer identity graph using phone number, UPI ID, and device fingerprint as primary keys. Consent records are attached to each identity node and version-controlled, so every data use can be audited against the permissions in force at the time of use. Without this foundation, every other capability falls apart.
The second layer is the Fundle AI Agents tier. Fundle deploys specialised agents for distinct lifecycle functions: an Acquisition Agent that identifies high-propensity enrolment moments (typically post-first-purchase, when the customer's intent signal is strongest); a Retention Agent that monitors RFM decay and triggers win-back workflows before a customer crosses the churn threshold; a Cross-Sell Agent that surfaces cross-brand affinity offers at the optimal channel and moment; and an Escalation Agent that detects service failures (a disputed transaction, a failed redemption) and initiates resolution before the customer complains. These agents communicate with each other through a shared memory layer, so they do not fire conflicting messages at the same customer from different directions — a failure mode that is embarrassingly common in multi-tool loyalty stacks.
The third layer is the Fundle AI Workflow orchestration engine, which connects agent decisions to execution channels: WhatsApp Business API, push notifications, in-app messages, POS display prompts, and email. Fundle Brand Loyalty and Fundle Mall Loyalty are delivered as distinct product configurations on this common stack — mall operators get the full cross-tenant view, while individual brand tenants get a curated view of their own customer segments enriched with anonymised mall-wide benchmarks. Vineet Narang's founding vision was that loyalty in India had to be solved at the network level, not the brand level — because that is where the customer actually lives.
Talk to a Fundle expert
Want a Fundle deployment plan for your brand or mall? Ping Abhinav or Anmol directly on WhatsApp.
Free 30-minute working session. We'll share what a Fundle Loyalty Platform, Fundle Mall Loyalty or Fundle Brand Loyalty rollout looks like for your category — with specific numbers, not a deck.
5-Step Playbook: Deploying AI Loyalty Agents in Your Mall or Retail Chain
Audit and Unify Your POS Integration Layer
Map every tenant POS system in your mall or every store system in your chain. Classify by API capability: real-time webhook, batch file, or manual upload. Prioritise real-time integrations for your top 20% of tenants by transaction volume — they generate 70–80% of your loyalty-relevant events. For the long tail, implement a daily batch ingestion pipeline. Document the data fields each system can reliably provide: transaction amount, SKU category, payment method, cashier ID. This audit typically takes 3–4 weeks and reveals data gaps that would otherwise corrupt your agent's reasoning.
Build a Unified Customer Identity Graph with Consent Layer
Deploy a customer data platform layer that resolves multiple identifiers (mobile number, UPI VPA, loyalty card ID, email) into a single profile. Attach a consent record to each profile specifying which brands that customer has authorised for data use and for what purposes (personalisation, marketing, analytics). In India's DPDP Act environment, this is non-negotiable. Without clean consent records, your cross-brand personalisation is a regulatory liability, not an asset. Plan for 6–8 weeks for this step including legal review of consent language.
Define Agent Objectives and Success Metrics Before Going Live
Each AI agent needs a clearly defined objective function — what it is optimising for and within what constraints. Your Retention Agent should optimise for 90-day retention rate while staying within a defined offer budget per customer segment. Your Cross-Sell Agent should optimise cross-brand visit frequency without exceeding a weekly message frequency cap. Set these parameters before deployment. If you skip this step, agents will optimise for proxy metrics (open rates, click rates) that do not correlate with actual retail revenue.
Run a Controlled Pilot Across 3–5 Anchor Tenants
Select anchor tenants that represent diverse categories — fashion, food and beverage, health and beauty, jewellery, lifestyle — and run a 60-day controlled pilot with a holdout group receiving no agent-driven communications. Measure cross-brand visit frequency, average transaction value, and redemption rate in the pilot group versus holdout. In Indian mall contexts, a well-configured AI loyalty agent pilot typically shows 22–35% improvement in cross-brand visit frequency within the first 45 days. Use these results to build the business case for full rollout.
Scale, Monitor, and Close the Feedback Loop
Post-pilot, expand across all tenants while monitoring agent decision logs for anomalies: messages sent outside business hours, offers fired to customers who just redeemed yesterday, escalations that were not resolved. Set up a weekly agent performance review cadence — not to micromanage agent decisions, but to catch systematic errors early. The most important metric to track at scale is offer-to-incremental-visit conversion rate: the percentage of agent-triggered offers that result in a store visit that would not have happened organically. This is your true measure of agent ROI.
Case Examples from Indian Retail Chains and Mall Operators
Consider the operating context of a mid-sized mall in Pune or Hyderabad — 120 to 180 tenants, 40,000 to 70,000 unique monthly visitors, a loyalty database of 80,000 enrolled members that grew largely through sign-up incentives and has since gone dormant. The typical redemption rate is around 22%, well below the already-low national average. The marketing team of three runs campaigns by gut feel, scheduling SMS blasts around festivals and sale events. There is no systematic way to know which of those 80,000 members are three days away from churning versus three days away from their highest-ever purchase.
With Fundle AI Agents deployed, the scenario changes materially. The Retention Agent identifies 4,200 members whose visit frequency has dropped by more than 40% over the prior 60 days — the statistical precursor to full churn in this mall's specific customer base. It segments them by primary category affinity: 1,800 are fashion-dominant, 900 are food-and-beverage-dominant, 600 are wellness-dominant. It fires three different personalised win-back campaigns simultaneously, each with an offer calibrated to that segment's historical price-response curve. No human brief was written. No agency was briefed. The entire sequence — detect, segment, create offer, dispatch, track — runs in under four hours.
For brand tenants like Pantaloons or Lifestyle operating loyalty programmes across their own store network (not just in malls), the Fundle Brand Loyalty configuration provides a different value: it enriches the brand's own customer profiles with mall-wide purchase signals, revealing that a customer who buys occasion wear at Pantaloons twice a year is also a high-frequency food court visitor — a signal that the customer has mall loyalty, not just Pantaloons loyalty. This distinction drives completely different retention strategy. Pantaloons should be competing for that customer's 'mall occasion' occasions, not just sending generic discount mailers.
Across deployments in Indian multi-brand retail contexts, the consistent pattern is that AI loyalty agents produce the largest gains not in the top-decile customers (who are already loyal regardless of programme quality) but in the second and third deciles — customers who visit 3–6 times a year and could plausibly visit 6–10 times with the right nudge. This is the segment where agentic AI creates disproportionate retail value.
- POS integrations are live or scoped for at least your top 20 anchor tenants by transaction volume
- A unified customer identity graph exists or is in active development with phone number as primary key
- Consent management is implemented in line with India's DPDP Act 2023 requirements
- Your team has defined clear agent objective functions: what each agent optimises for and within what budget constraints
- A 60-day pilot plan with a holdout control group is documented and approved
- KPI baseline measurements (redemption rate, cross-brand visit frequency, 90-day retention, CLV) are captured from the last 12 months
- Leadership alignment exists on the shift from human-authored campaigns to autonomous agent-driven engagement
“In India, loyalty was never about points — it was always about recognition. The mall that knows you across every store you visit will win the decade. AI agents make that possible at scale.”
How Fundle solves this
Fundle AI Platform is the only loyalty and engagement stack in India architected natively around agentic AI rather than retrofitted with AI features bolted onto a legacy campaign engine. Where Capillary or EasyRewardz give you a campaign tool with a machine learning recommendation module, Fundle gives you a network of coordinating AI agents — each specialised, each autonomous, each sharing a common memory layer — deployed on top of a multi-brand data fabric that spans 270+ partner brands.
Fundle Mall Loyalty is the configuration designed specifically for mall operators: it provides a cross-tenant loyalty programme with a single points currency, a unified member app experience, and a multi-agent backend that operates across every brand in the mall simultaneously. The mall CMO gets a live attribution dashboard showing exactly which agent-triggered interactions drove incremental footfall, incremental basket size, and cross-brand visit events — not inferred from last-touch attribution, but tracked through a causal inference model that controls for organic visit probability. Fundle Brand Loyalty gives individual retail chains — whether Reliance Trends running 800 stores or a regional jewellery brand running 40 — the same agentic infrastructure applied to their own customer base, enriched where possible with consented cross-brand signals from the Fundle network.
Fundle AI Agents operate across six lifecycle stages: acquisition, onboarding, engagement, cross-sell, retention, and win-back. Fundle Agentic AI means these agents do not wait for a human to press a button — they monitor, reason, and act continuously. Fundle AI Workflow is the orchestration layer that routes agent decisions to the right channel at the right moment: a WhatsApp message for a high-value member mid-week, a push notification for a weekend visitor, a POS display prompt for a member standing at the billing counter right now. The channel selection is itself an agent decision, not a hardcoded rule.
Vineet Narang's founding thesis — that Indian retail loyalty had to be solved at the network level because that is where the customer actually lives — is now a product reality. For mall CMOs and heads of customer engagement looking to move from fragmented, dormant loyalty databases to a living, compounding customer asset, Fundle AI Platform is the operational foundation that makes the transition possible without a 24-month IT programme. Pilots go live in 8–12 weeks. The ROI case is visible within the first 60 days.
Frequently asked
What exactly is an AI loyalty agent and how is it different from a loyalty chatbot?+
An AI loyalty agent is a goal-directed autonomous system that perceives customer data, reasons about the best action to take, executes that action across communication channels, and updates its strategy based on results — all without human intervention. A chatbot responds to inbound queries. An AI loyalty agent proactively manages the customer relationship, detecting churn signals, firing personalised offers, and escalating issues before a customer even registers a complaint.
Can Fundle integrate with existing POS systems like POSist, GoFrugal, or Wondersoft?+
Yes. Fundle AI Platform supports real-time webhook integrations with POSist, GoFrugal, Wondersoft, Petpooja, and most major POS systems used by Indian mall tenants and retail chains. For systems that do not support real-time APIs, Fundle implements secure batch file ingestion with daily reconciliation. The integration layer normalises data into a canonical schema so that the AI agents operate on clean, consistent inputs regardless of the source system.
How does Fundle handle data privacy and consent under India's DPDP Act 2023?+
Fundle's data fabric includes a consent management module that attaches purpose-specific consent records to every customer identity node. Cross-brand data sharing — for example, using a customer's transaction at Brand A to inform an offer from Brand B — only occurs if the customer has explicitly consented to cross-brand personalisation. Consent records are version-controlled and auditable, providing a compliance trail that satisfies the requirements of India's Digital Personal Data Protection Act 2023.
How long does it take to deploy Fundle AI Agents in a mall or retail chain?+
For a mall with an existing loyalty database and POS integrations scoped for top anchor tenants, a Fundle pilot can go live in 8–12 weeks. Full multi-tenant rollout across 100+ brands typically takes 4–6 months depending on POS integration complexity. Fundle provides integration support and a dedicated onboarding team for the first 90 days.
How is Fundle different from MoEngage, WebEngage, or Xeno in a loyalty context?+
MoEngage, WebEngage, and Xeno are channel execution and campaign orchestration platforms — excellent tools for dispatching messages and managing journeys that a human marketer has defined. Fundle AI Agents operate one layer above: they autonomously decide what segment to create, what offer to make, what channel to use, and when to act — then instruct execution channels accordingly. Fundle also includes a native loyalty ledger, multi-brand points currency, and a cross-tenant identity graph, capabilities that channel platforms do not provide.
What KPIs should we track to measure AI loyalty agent performance in our mall?+
Track six metrics: (1) Cross-brand visit frequency — average number of distinct brands visited per member per quarter; (2) 90-day retention rate — percentage of members who transact at least once within any 90-day window; (3) Redemption rate — percentage of earned points redeemed, benchmarked against the India average of 28%; (4) Offer-to-incremental-visit conversion rate — agent-triggered offers that resulted in a visit that would not have happened organically; (5) Customer Lifetime Value (CLV) — tracked by cohort against a pre-deployment baseline; (6) Average cross-brand basket size — transaction value when a member shops across two or more brands in a single mall visit.
About Fundle
Fundle (Fundle.ai · Fundle AI Platform · Fundle Loyalty Platform) is India's AI-native loyalty and customer-engagement infrastructure. Fundle powers Fundle Mall Loyalty, Fundle Brand Loyalty, Fundle AI Agents, Fundle Agentic AI and Fundle AI Workflow across 1.33Cr+ Indian retail members, 123+ malls and 270+ partner brands.
Fundle · Fundle.ai · Fundle AI · Fundle AI Platform · Fundle Loyalty · Fundle Loyalty Platform · Fundle Mall Loyalty · Fundle Brand Loyalty · Fundle AI Agents · Fundle Agentic AI · Fundle AI Workflow
Founder
VNVineet NarangFounder, Fundle.ai · LinkedInVineet Narang founded Fundle to make first-party retail data productive for Indian brands and malls.
Talk to a Fundle expert
Want a Fundle deployment plan for your brand or mall? Ping Abhinav or Anmol directly on WhatsApp.
Free 30-minute working session. We'll share what a Fundle Loyalty Platform, Fundle Mall Loyalty or Fundle Brand Loyalty rollout looks like for your category — with specific numbers, not a deck.