Handle order tracking, returns, and delivery inquiries without hiring a single seasonal agent — the same agent answers from phone, web, and your mobile app.
Intelina Voice· Phone
Order tracking, returns and delivery changes on the phone through peak season without seasonal hiring.
Intelina Desk· Web chat
Storefront widget answering 'where is my order' from your OMS in real time, opening a ticket only for exceptions.
50K+ calls
Peak Capacity
0 agents
Zero Hiring
90%+
Target Resolution
Figures on this page are planning targets and typical volume distributions for the sector, not measured results from a named customer. We size them against your own numbers during the demo.
The Seasonal Scaling Challenge
10x call volume spikes during holidays
Hiring and training temporary agents
Inconsistent service quality
Order tracking system fragmentation
Customer
Menu or queue
Dead end
Contact mix
Where the volume actually goes.
Before deciding what to automate, it helps to see what people are actually contacting you about — and which module is the right home for each slice.
38%Where is my orderIntelina Desk
TodayAgent copies a tracking number
21%Returns, exchanges and refundsBoth modules
TodayMulti-step manual process
14%Delivery changes and reschedulesIntelina Voice
TodayPhone queue during peak
12%Product, stock and promo questionsIntelina Desk
TodayPre-sale chat, often unanswered
9%Payment and checkout failuresBoth modules
TodayCart abandoned before anyone replies
6%Complaints needing a humanBoth modules
TodayBuried in a shared inbox
The approach
Elastic capacity that scales with demand
01
Scales instantly
Volume spikes are absorbed automatically — no recruiting, onboarding, or overtime during peak season.
02
Knows every order
Connected to your OMS and carriers, the agent gives live tracking, ETAs, and delivery updates.
03
Processes returns end to end
Initiates returns, issues labels, and triggers refunds in the conversation — consistent every time.
04
Consistent on every channel
The same agent and context answer whether a shopper calls, chats on the site, or taps in the app.
What the agent handles
Real-time order tracking & ETAs
Returns, exchanges & refund initiation
Promo, stock & product questions
Delivery rescheduling & address changes
Playbooks
What each module actually does here.
Not capabilities in the abstract — the specific flows, in order, with the primitives each one relies on.
Intelina Voice
Phone · AI voice agents + customer CRM
Order status without a reference number
TriggerCaller asks where their order is
1Caller ID resolves to the customer record before they finish the sentence
2The agent calls your order system for live status and carrier ETA
3It reads back the actual delivery window rather than a generic range
read_customerhttp_webhook → OMS
Returns started and finished on the call
TriggerCaller wants to return or exchange an item
1The agent confirms the item from the order history
2It opens the return in your system and triggers the label
3The reference is spoken back and texted after the call
http_webhook → returnspost-call SMS
Peak volume without peak headcount
TriggerTen times the normal call volume in a holiday week
1The audio bridge and the API scale independently of each other
2Every call gets the same persona, knowledge and tools as the first
3Nothing depends on training a seasonal cohort
horizontal scalingshared agent configuration
Intelina Desk
Web chat · AI chat + ticketing tool
Tracking answered inside the storefront
TriggerSigned-in shopper opens the widget
1The widget knows who they are from your own session endpoint
2A registered endpoint returns their live order list
3The bot answers before the shopper types an order number
signed-in user detectiondata endpoints
Refund policy that quotes your policy
TriggerQuestion about returns windows or eligibility
1The knowledge base is searched over the policy documents you uploaded
2The answer stays inside what those documents actually say
3Edge cases the documents do not cover surface as knowledge gaps
knowledge baseknowledge gaps
Only the exceptions reach a human
TriggerDamage, dispute, or a request for a person
1The chat becomes a ticket in the right queue with the full thread
2SLA tracking starts immediately so peak-season backlogs stay visible
3The agent sees the order data the bot already fetched
escalationqueuesSLA tracking
Integrations
What it plugs into.
Every connection is something you register in the dashboard — no bespoke integration work on our side.
Order management system
Registered data endpoint plus an HTTP webhook tool for writes
Carrier and tracking APIs
Data endpoint with a field whitelist on the response
Returns and refunds platform
HTTP webhook tool with idempotency on submit
Storefront
One widget script tag, themed to your brand
Support inbox and queues
Ticketing workspace with per-team routing
Guardrails
What it refuses to do.
The limits that matter in this sector, enforced by configuration rather than by prompt wording.
No invented delivery dates
Dates come from a live tool call. If the call fails, the agent says so rather than estimating.
Refunds are idempotent
A retried refund replays the original result instead of issuing a second one.
Peak-season isolation
Every project's queries, caches and vector searches are scoped, so one tenant's surge cannot reach another's data.
Approvals on goodwill credits
Discretionary refunds and credits can be held for a supervisor rather than granted by the AI.
Rollout
How this actually goes live.
Read-only first, writes behind approvals, then widened once the record justifies it. No big-bang cutover.
01Day 1
Drop in the widget
One script tag on the storefront, themed to your colours and logo. No rebuild, no SDK.
02Days 2–5
Point it at orders
Register the order and tracking endpoints, whitelist the fields, and test each one against a sample user.
03Week 2
Load the policies
Upload returns, shipping and promo documents so answers are quoted rather than improvised.
04Week 3
Open the phone line
Map a number, import customer records, and turn on the order-status and returns tools for voice.
05Before peak
Rehearse the surge
Load-test ahead of the season, set queue routing and SLA thresholds, and decide which actions stay behind approvals.
Measurement
What to hold it accountable to.
Each target maps to something the product already reports, so nobody has to build a spreadsheet to find out whether it worked.
Metrics, current typical state, target, and where each is reported
Metric
Typically today
Target
Reported by
Self-service resolution
Most contacts reach an agent
90%+ of order enquiries
Ticket volume by category and escalation rate
Seasonal hires needed
A cohort every peak
None
Contacts handled per human agent
First response time at peak
Degrades with volume
Flat regardless of volume
Project analytics, first-response metric
Repeat contacts per order
Multiple 'where is it' checks
One, or none
Tickets grouped by order reference
Projected outcomes
What success looks like
50K+
Peak calls handled
Holiday surges absorbed with zero new hires.
90%+
Self-service resolution
Most inquiries resolved without an agent.
0
Seasonal agents needed
Capacity scales on demand, not on headcount.
Black-Friday-scale volume, answered instantly and consistently — without a hiring scramble.
E-Commerce questions
What buyers in this sector ask first.
Chat is same-day: a script tag, a knowledge upload, and one registered order endpoint. Voice takes longer because a number has to be mapped and customer records imported — plan a short onboarding rather than a project.
See this running on your own e-commerce data.
Bring one line or one queue and your real numbers. We will size the plan against them on the call rather than quoting the targets on this page back at you.