2026
Milo x loto.lt : AI-Powered eCommerce Analytics Case Study
In modern eCommerce and digital lottery platforms, speed is competitive advantage. Acquisition funnels, conversion behaviour, and customer top-up patterns must be analysed in real time - not weeks later. For Perlas Network, Lithuania’s leading lottery distribution operator, traditional business intelligence workflows created delays that slowed optimisation. This case study explores how Perlas Network implemented AI-powered conversational analytics with Milo to transform reporting cycles from weeks into minutes.
Faustas Rimkevičius
Growth Marketing
About Perlas Network (Lithuania’s Leading Lottery Distribution Platform)
Perlas Network operates one of Lithuania’s largest lottery distribution networks via both physical retail and its online platform (loto.lt).
Industry: Lottery Distribution & eCommerce
Role Interviewed: Head of eCommerce
Technology Environment: Internal analytics systems + replaced PowerBI with Milo
Access Channels with Milo: Conversational interface + live dashboards
Core Focus Areas:
Customer acquisition funnels
Sales conversion optimisation
Top-up behaviour analytics
Customer segmentation & cohort analysis
Retention & engagement tracking
Beyond ticket sales, Perlas delivers the emotional experience of possibility and winning. Maintaining that experience online requires deep, real-time insight into customer behaviour.
Watch the full video of this case study here:
The Core Problem: Behavioral Insight Took Weeks
Before Milo, and during their reliance on PowerBI, understanding customer behavior patterns required multiple manual iterations.
As Erika explains:
“To understand customer behavioral patterns or to get any kind of numbers usually took time or even weeks. Often with a number of iterations with our data analyst.”
The workflow looked like this:
Define a question (e.g., funnel drop-off, churn spike, campaign ROI)
Submit a request to a data analyst
Wait for data extraction and PowerBI report updates
Review results
Request clarification or segmentation
Repeat cycle
This created three structural bottlenecks.
1. Slow Behavioral Pattern Analysis
Customer behavior in eCommerce changes quickly. According to McKinsey & Company, companies that leverage real-time behavioral analytics significantly outperform competitors in customer engagement and personalization.
When behavioral insight takes weeks, response windows close.
2. Iterative Friction with Analytics Teams
Every segmentation, cohort filter, or funnel breakdown required analyst support to manipulate PowerBI. This limited experimentation speed and delayed optimization cycles.
Research from Gartner highlights that decision intelligence capabilities reduce operational latency and improve business responsiveness.
Without real-time access, teams default to retrospective reporting instead of proactive optimization.
3. Focus on Collecting Numbers Instead of Acting
Perhaps the most important impact was cultural.
Time was spent retrieving data rather than executing improvements.
As Erika describes:
“Team doesn’t need to look for the numbers. Numbers are there for them so they can be focused on what truly matters.”
Why Milo: From Data Retrieval to Behavioral Intelligence
Milo introduced a structural shift, replacing the rigid constraints of PowerBI with a conversational analytics layer:
From requesting numbers → to understanding drivers.
From waiting → to acting.
Instead of dashboards-first analytics, Milo enables:
Instant behavioral queries
Funnel monitoring
Acquisition & retention diagnostics
Churn detection
Cross-metric analysis
Erika and her team now use Milo daily.
“We use Milo every day for any occasion.”
This daily adoption is critical. According to Forrester Research, organizations that embed AI into operational workflows (rather than using it as a reporting add-on) see the strongest productivity gains.
Milo became embedded in daily decision-making.
Solution in Action: Monitoring Funnels, Campaigns & Churn in Real Time
Use Case 1: Sales Funnel Optimization
The team continuously monitors:
Landing page performance
Registration conversion rates
Deposit conversion
Drop-off points
Funnel velocity
When friction appears, they adjust immediately.
Previously, funnel diagnostics required iterative PowerBI reporting cycles. Now they are instant.
Use Case 2: Acquisition & Retention Campaign Performance
The team tracks:
Cost per acquisition (CPA)
Customer lifetime value (LTV)
Retention curve movement
Campaign-specific revenue lift
Cohort behavior over time
If performance shifts, campaigns are adjusted in real time.
“We can change the campaigns fast, we can adjust funnels fast.”
Speed transforms marketing economics.
Use Case 3: Churn Monitoring & Behavioral Signals
Churn compounds silently in subscription and lottery ecosystems.
With Milo, the team:
Identifies churn spikes
Analyzes behavioral drivers
Compares against historical baselines
Tests corrective interventions
Instead of discovering churn in monthly PowerBI reviews, they act during daily monitoring.
The Impact: Confidence + Speed
The most significant outcome wasn’t just operational.
It was psychological.
“It brought the confidence to me and my team because we understand the numbers, we understand how they work with each other and how they influence the result in general.”
Milo delivered:
Instant answers
Unified dashboards
Cross-metric clarity
Behavioral transparency
Decision confidence
Erika emphasizes two core strengths:
“The best part of Milo is speed of answers and probably the dashboards.”
Dashboards provide visibility.
AI provides interpretation.
Together, they eliminate hesitation.
Before vs After: eCommerce Analytics Transformation
Before (PowerBI Era) | After with Milo |
Weeks to understand behavioral patterns | Instant behavioral insights |
Multiple analyst iterations for PowerBI | Daily funnel monitoring |
Manual segmentation requests | Live acquisition & retention analysis |
Delayed campaign adjustments | Immediate churn diagnostics |
Focus on data retrieval | Unified dashboards |
Reactive optimization | Focus on action, not collection |
This was not incremental improvement.
It was operational acceleration.
Cultural Shift: From Reporting to Action
Perhaps the most important change was focus.
Before: Energy went into collecting numbers via manual PowerBI workflows.
After: Energy goes into acting on insight.
“Numbers are there for them so they can focus on what truly matters.”
This shift reflects a broader transformation toward AI-powered decision intelligence across industries.
Frequently Asked Questions
What is AI-powered eCommerce analytics?
AI-powered eCommerce analytics uses artificial intelligence to analyse customer acquisition, conversion funnels, top-up behaviour, and retention in real time. Instead of relying on static dashboards, teams can ask questions in natural language and receive instant, actionable insights that support faster decision-making and revenue optimisation.
How does conversational business intelligence work?
Conversational business intelligence allows users to ask data questions in plain language. The AI translates those questions into data queries, analyses connected systems, detects patterns, and returns clear insights within minutes. This removes the need for manual report building and reduces dependency on BI teams.
How can AI analytics improve conversion rates?
AI analytics improves conversion rates by identifying funnel drop-offs, segmenting customer behaviour, detecting purchasing patterns, and highlighting optimisation opportunities in real time. Faster insight allows teams to test changes quickly, refine acquisition strategies, and increase customer lifetime value.
What is decision intelligence in eCommerce?
Decision intelligence in eCommerce refers to AI-driven systems that turn raw data into actionable recommendations. Instead of reviewing static reports, teams receive contextual explanations, performance comparisons, and suggested next steps — enabling faster, more confident business decisions.
How is AI analytics different from traditional BI tools?
Traditional BI tools rely on pre-built dashboards and manual report creation. AI analytics platforms use natural language processing and automated pattern detection to generate real-time insights instantly. This reduces reporting delays and allows business teams to self-serve answers without technical queries.
Is AI-powered business intelligence secure?
Enterprise AI analytics platforms use encryption, role-based access control (RBAC), and secure data connections to protect sensitive information. Access permissions ensure users only see authorised data while maintaining real-time insight capabilities across the organisation.
