Retail & eCommerce

From 'where is my order?' to 'here's what you'll love next'

Agents that resolve order status, returns, and delivery coordination end-to-end — connected to inventory, logistics, payments, and loyalty — and that turn service moments into revenue moments.

See the pilot
The executive problem

Retail communications are dominated by a few high-volume, data-backed intents — order status, returns, delivery changes — that customers want answered instantly, in their language, at 10pm during a sale.

Every service contact is also a revenue moment: a recovered cart, a loyalty nudge, a cross-sell. Fragmented bots and outsourced queues capture none of it.

Peak-season economics break human-only models: hiring and training seasonal agents for a six-week spike is the most expensive way to answer 'where is my order?'.

Why now
  • WISMO (where-is-my-order) and returns queries routinely exceed half of all support contacts
  • Conversational commerce on WhatsApp and RCS is mainstream in India and emerging markets
  • Marketplace seller ecosystems need scalable support that human teams can't economically provide
  • Festival / holiday peaks force a choice between poor service and bloated seasonal cost
50%+
of support contacts are order-status and returns queries in most e-commerce operations
operator benchmarks; validate per customer
~70%
of shopping carts are abandoned — recoverable with timely, contextual outreach
Baymard Institute meta-analysis
24/7
coverage in every language, at flat marginal cost during peaks
platform capability
Priority use cases

Ranked by value, feasibility, and speed

Start where value and feasibility intersect. Tags mark the lighthouse, the fastest path to production, and the plays partners lead.

Use caseValueFeasibilitySpeedNote
Order status & delivery coordination (WISMO)
Lighthouse
Highest volume, clean OMS/logistics APIs, instant CSAT impact
Returns & exchanges end-to-end
Highest volume
Policy-bounded actions: label generation, refund initiation
Abandoned-cart & drop-off recovery
Fastest to production
Direct revenue attribution; consent-governed outreach
Product discovery & guided selling
Executive value
Catalog grounding; measurable conversion lift
Delivery-failure resolution & address fixes
RTO reduction is pure margin in COD-heavy markets
Loyalty engagement & win-back
Segmented outreach from CDP triggers
Marketplace seller support
Partner-led
B2B intents: payouts, listings, penalties
Post-purchase care & setup help
Warranty, installation booking, how-to
Store associate assist
Inventory lookup, endless-aisle in store
Promotional & campaign outreach
Governed frequency caps; suppression rules
Lighthouse journey

WISMO and delivery coordination, before and after

Today
  1. 1Customer digs through email for a tracking link, then calls; IVR offers no order context
  2. 2Agent asks for order ID, re-authenticates, toggles OMS + logistics screens
  3. 3Delivery change requires a ticket to another team; customer calls back twice
  4. 4Failed deliveries become RTO losses; nobody proactively tells the customer
  5. 5Peak season: 45-minute queues or an expensive seasonal bench
With the agentic layer
  1. 1Customer messages on WhatsApp or calls; agent already knows their recent order
  2. 2Live status from OMS and the logistics partner, in the customer's language
  3. 3Agent reschedules delivery or fixes the address directly in the logistics system
  4. 4Proactive notification when a delivery is at risk — before the customer asks
  5. 5Peaks absorbed at flat marginal cost; humans handle disputes and exceptions
Watch the transformation

The same interaction, two worlds — racing live

The loop runs itself: the conventional journey stalls, backtracks, and drags while the agentic one flows straight through — grounding, acting, and resolving with a human approval exactly where policy demands one. Pause anytime and click any node to explore that step.

Scenario: A customer's cash-on-delivery order shows 'out for delivery' for the second day
Today
same interaction, two worlds
Agentic layer
Time to resolution
26+ min + 1–2 day ticket
under 2 minutes, in-channel
Assumption
Contacts per incident
3–4 across channels
1 proactive thread
Platform capability
Cost per contact
$13.50 median assisted
$1.84 median self-service
Benchmark
Cart abandonment context
~70% of carts abandoned
recoverable via consent-based outreach
Benchmark
RTO reduction
—
measured in pilot
Assumption
Sources & assumptions
  • · Time to resolution: Conventional = queue + handle + ticket SLA (typical operator pattern, assumption); agentic = platform design
  • · Contacts per incident: Process design: exception detected before the customer calls
  • · Cost per contact: Gartner customer service cost benchmarks, 2024
  • · Cart abandonment context: Baymard Institute meta-analysis; recovery rate measured in pilot
  • · RTO reduction: COD RTO economics vary widely by category; baseline yours before projecting
  • · Gartner, customer service cost benchmarks (2024): $13.50 median assisted vs $1.84 self-service per contact
  • · McKinsey, digital-first collections research: 20–25% NPL reduction among leaders; up to 40% opex reduction with gen AI
  • · Baymard Institute: ~70% average cart abandonment (meta-analysis)
  • · IAMAI–Kantar via IBEF (2025): 900M+ Indian internet users; 98% consume Indic-language content
  • · LeadSquared and vendor funnel studies: 78% of students choose the first institution to respond (directional, vendor data)
  • · HDI / ITSM operator benchmarks: $15–25 per L1 ticket; 40–60% of L1 volume is resets/status (validate per customer)
  • · Conventional-flow wait times and volumes are typical operator patterns — assumptions to replace with the customer's own baseline
  • · Agentic-flow behaviors (context retention, 100% logging, in-line policy checks) are platform capabilities, not projections
One integrated solution — clear ownership
Gemini Enterprise Agent Platform

Agent Platform orchestrates exception detection, conversation, policy reasoning, and write-actions into OMS/3PL systems.

Customer Engagement Suite (GECX)

If the retailer runs a CC platform, human escalations arrive there context-complete; GECX leads if the engagement is CC modernization.

Tilicho Labs — communications & implementation

Tilicho Labs runs the channel and voice layer and builds the 3PL/OMS integrations and category-specific configurations; its platform absorbs peak volumes on usage-based pricing.

Human oversight

Ops leads approve above-tier goodwill; disputes and fraud flags always route to people.

Customer & employee impact

Customers: Told about problems before they notice, given real options in one thread, compensated fairly — festival peaks feel like off-peak.

Employees: Service teams stop being human middleware between OMS and 3PL portals; they handle disputes and exceptions with live context.

Value drivers · differentiators
RTO avoidance on COD ordersPeak absorbed at flat marginal costWISMO removed from human queuesProactive outreach converts service into retentionTicket elimination for routine logistics actionsOne agent layer across WhatsApp, voice, web, and appReal-time grounding in OMS + logistics truthPolicy-tiered goodwill with human approval above thresholdPeak elasticity on Google infrastructure
Operating-model comparison

Conventional vs point solutions vs agentic

Thirteen dimensions where the operating models differ — from experience and speed to governance and measurement. Point solutions improve one channel; the agentic layer changes how the whole enterprise communicates.

Conventional: Queue, IVR maze, repeat yourself at every step

Point solutions: One channel improves; experience breaks at its edges

Agentic: Recognized, understood, and resolved in the customer's language on any channel

Who you'll talk to

Every stakeholder, one coherent story

CEO

Cares about: NPS and cost per order

Wants: Service cost decoupled from order growth

Will object: Bots have burned us before

Proof that works: Containment + CSAT on one intent in 8 weeks

“What does answering 'where is my order?' cost you per order today — and at peak?”
CIO / CTO

Cares about: Integration sprawl across OMS, WMS, logistics

Wants: One governed action layer over existing APIs

Will object: Our stack is a patchwork

Proof that works: Connector inventory + reference architecture session

“The agent layer sits over your existing APIs — no rip-and-replace.”
Customer Service leader

Cares about: Queue times, seasonal hiring

Wants: 60–80% containment on top intents

Will object: Escalations will just get harder

Proof that works: Warm-handoff demo with full context transfer

“If WISMO disappeared from your queues, what would your team do with the time?”
CMO / Growth

Cares about: Cart recovery, retention

Wants: Conversational outreach with attribution

Will object: Outreach fatigue and unsubscribes

Proof that works: Frequency-capped, consent-governed campaign design

“Your cart-recovery emails get 2% response. What does a two-way conversation get?”
Operations leader

Cares about: RTO losses and delivery failure

Wants: Proactive coordination cuts failed deliveries

Will object: Logistics partners won't integrate

Proof that works: Partner-built connectors to major 3PLs

“Every failed COD delivery is pure margin loss. What's your RTO rate?”
Architecture

The same layers, grounded in your systems

Channels and industry connectors are partner-built; the agent platform and models are Google Cloud; your systems of record and policies remain yours.

Gemini Enterprise Agent Platform

Enterprise Intelligence & Agents

Google Cloud provides the intelligence: agents that reason over policies, history, and enterprise data, coordinate specialized agents, and execute governed actions — running in the customer's Google Cloud environment.

Gemini reasoning & intentEnterprise grounding with citationsDomain & workflow agentsMulti-agent orchestrationEnterprise-system actionsPolicy enforcement & model routing
1Contact

Interaction arrives on any channel — or the enterprise initiates outreach

2Identify

Identity and permissions resolved; consent and policy windows checked

3Understand

Intent understood in the customer's language, with full context retrieved

4Reason

Agent reasons over policies, history, and enterprise data

5Act

Governed actions executed in systems of record — payments, bookings, tickets, updates

6Escalate

Humans brought in on sentiment, complexity, or policy triggers — with full context

7Record

Outcome written back; 100% of the interaction logged and auditable

8Improve

Every interaction evaluated; policies, prompts, and workflows refined

Economics

Model it with your numbers

Defaults reflect typical retail & ecommerce interaction profiles — replace them with yours. Every formula is shown; the pilot proves the containment rate before any scale decision.

Your inputs
800K
3 min
$0.35
$0.06
70%
$250K
Modelled impact
$545K
monthly cost savings
0.4 mo
payback period
$19.8M
three-year net value
560K
interactions resolved by agents / month
Sensitivity — containment ±15 points
$469K
$620K
monthly savings range across containment scenarios
Assumptions & formulas (no hidden numbers)
  • Baseline cost = interactions × minutes × human $/min
  • Contained interactions cost AI $/min only; escalated interactions incur AI cost + 60% of human handle time (triage saves the rest)
  • Revenue uplift = contained volume × $1.2/resolved × 2% incremental outcome lift (conservative default — replace with pilot data)
  • Containment is the pilot's job to prove — the sensitivity band shows the stakes. These are modelling defaults, not claims.
Security & regulatory

Designed for your regulatory reality

Consent and opt-out management for proactive outreach, per channel
Payment actions bounded: refunds within policy tiers only, above-tier to humans
PII minimization in prompts and logs; regional data residency
Marketplace regulations: seller communications and penalty disputes documented
The pilot

Narrow scope. Real traffic. Explicit thresholds.

Scope · 6–10 weeks

WISMO + delivery coordination on WhatsApp and web chat, two languages, one region

Integration boundary
Order management system (read)Logistics partner API (read/write reschedule)WhatsApp BSP via Tilicho Labs
Success is measured as
  • Containment on WISMO intents
  • CSAT vs human baseline
  • RTO / failed-delivery delta
  • Cost per contact
  • Escalation quality score
Expansion roadmap

One pilot, then the platform

1
Returns automation
2
Cart recovery
3
Guided selling
4
Seller support
5
Store associate assist
Common questions

Asked in every meeting