
One identity layer across every Reliance retail business — and the agentic marketing layer being built on top of it. UCP is live across 100% of Reliance Retail formats with 400M+ identities and 1.5M daily RPOS orders. Marketing OS — three agentic cockpits (Marketing Intelligence · Brand Health · AI-Native UCP) — climbing to Agentic L4.
Powered by the Marketing Agent → · 11-agent topology · Meta-Orchestrator + 9 Domain Agents + Compliance · /agents/ catalog.
L-rungs reference the L0–L5 autonomy framework. UCP foundation sits at L3; Marketing OS cockpits target L4.
Each layer earns the next. Identity made unified data possible. Unified data made segmentation and activation possible. Activation at scale is what makes the agentic layer worth building.

DPDP-compliant identity with consent capture, unified address book, retail SSO, and API-first integration for any platform regardless of design system.
Web engagement, offline store visits, retail KPIs in one 360° view per customer at RRL, Format, and Business level.
From audience to landed message to attributed conversion — closed loop, governed by cooldowns and volume caps across 200+ formats.
Three cockpits answering the three questions every vertical leader asks: are we tracking to plan, is marketing pulling its weight, do consumers know us and prefer us.
Marketing OS sits on top of UCP. Three cockpits today — each answers a different question every vertical leader asks. Each carries an L-rung pill for current state and target. Reference: L0–L5 autonomy framework.
The cockpit for vertical leaders — AOP-vs-actuals, marketing efficiency, customer mix, and platform health in one view. Replaces the WBR-decks-plus-finance-trackers-plus-MarTech-dashboards stitch leaders did each month. Answers three questions in sequence: are we tracking to plan, is marketing spend pulling its weight, what is the data telling us we are missing.
Single scrollable canvas grouped into three bands — Business Outcomes (Net Revenue · Orders · Customers · ABV with AOP achievement), Marketing (Total Spend · A2S% · PM · Blended CAC · LTV/CAC), Platform & Brand Health (MAU · DAU · Sessions · NPS). Every KPI carries an AOP-anchored comparison, not just a YoY delta.
Vertical narrative · outcome → cause → diagnosis. Revenue Story (actuals vs annual plan, 13-month trends, source splits New/Repeat/Winback, retention). Marketing Effectiveness (A2S, ROAS, Blended + PM CAC, LTV/CAC, approved-vs-actual budget, channel breakdown ATL/BTL/PM). AI Diagnostic with colour-coded insight cards · Red material misses · Amber watchpoints · Green positive · Blue neutral.
Conversational layer over the same dataset. "Why did ABV drop in Feb?" / "Show me Tira CAC trend vs Ajio." Charted, sourced answers without writing SQL or pulling reports.
What makes it agentic. Plan-aware by default · grounded AI diagnostics that reference visible numbers · same shape across every vertical (Ajio · JioMart · Tira · Reliance Digital · F&L). Switch the vertical, get the same cockpit.
Where Marketing Intelligence answers "are we hitting the plan?", Brand Health answers "do consumers know us, prefer us, use us, trust us — and how do we compare to the competition?" One brand with every angle — funnel · imagery · usage · zonal · purchase behaviour · creative recall — in a single experience.
Side-by-side funnel · anchor brand (left) vs chosen competitor (right). Five stages: Total Awareness → Consideration (T2B) → L1Y Usage → L3M Usage → MOUB. Each stage shows absolute score, ±pp delta vs prior wave, relative % change, conversion ratios. Competitor selector swaps the comparison brand without leaving the view (Blinkit · Zepto · Swiggy Instamart · Flipkart Minutes · Amazon Fresh · Bigbasket Now).
Vertical narrative · Brand Imagery (20 attributes across functional / commercial / emotional, gaps table sorted by magnitude, top-5 movers QoQ). Usage Funnel (trial → habitual, vs best-in-class named explicitly). Purchase Behaviour (Kirana · Quick-commerce · Modern Trade). Zonal Performance (North · South · East · West). Ad Testing Results (creative-level masked brand-recall). AI Analysis · 4 pillars · Perception / CX / Growth / Marketing.
"How has JioMart's MOUB tracked vs Blinkit over the last 4 waves?" "Which imagery attributes improved the most in South zone?" Charted, sourced answers from the brand-tracker dataset.
What makes it agentic. Gap-first rather than score-first design (the delta matters more than the absolute in mature categories) · side-by-side competitor comparison built into every metric · creative recall closing the loop from campaign spend to brand equity.
A complete CDP in a single chat interface. The target for Marketing OS — every step of the marketing workflow (data → cohort → campaign → creative → execution → retrospective) happens conversationally, with AI models powering each step. Demo · youtu.be/9KCFnV6kXEU.
What makes it L4. No dashboards. No tool-switching. Governance guardrails built in. The reviewer can ask the question and get a charted, sourced answer that ends in a launched campaign.
10 RCS campaigns · 11 - Apr - 2026 → 19 - Apr - 2026 · UCP's first large-scale conversion-tracked campaign for RD. End-to-end · audience segmentation through RCS delivery to offline POS conversion tracking, all powered by UCP.
Top brands · Samsung ₹3.98 Cr · Apple ₹3.19 Cr · LG ₹2.59 Cr. Top SKUs · Voltas + LG split ACs (seasonal) · iPhone 17 Pro Max. Geography · South 44.5% (₹9.07 Cr), West 20.3%, East 19.4%, North 15.9%. Saturday 19 - Apr alone generated ₹4.08 Cr · 20% of total revenue in one day. Replicable playbook for electronics-retail RCS campaigns now established.











