>
Poornachandra
Kongara
@poornachandra-kongara
Descriptive Analysis
(What happened?)
Descriptive analysis is where every
analytics project begins.
It focuses on summarizing raw data into clear
business numbers so stakeholders can immediately
understand the current situation.
Instead of looking at thousands of rows,
you convert data into:
KPIs (Revenue, Orders, Users, Churn)
totals, averages, counts
summary tables and dashboard cards
Use cases:
Monthly sales report
Website traffic summary
Customer growth tracking
What it delivers:
A clean picture of business performance — what happened,
how much happened, and where it happened.
>
Poornachandra
Kongara
@poornachandra-kongara
Trend Analysis
(What is changing over time?)
Trend analysis helps you understand the
direction in which the business is moving.
It answers questions like:
Are we growing or declining?
Is the change gradual or sudden?
Is there seasonality (weekends, festivals, month-end
spikes)?
Did something break or improve recently?
This is where analysts identify patterns hidden across
time.
Use cases:
Revenue trend by month
Churn trend after pricing change
Daily active users trend
What it delivers:
Clear visibility into time-based performance — making it
easy to spot early warning signs and growth opportunities.
>
Poornachandra
Kongara
@poornachandra-kongara
Comparative Analysis
(What’s better/worse?)
Comparative analysis is used when businesses
want clarity on performance differences.
Instead of saying “sales increased,” you show:
which region improved the most
which product performed worst
which time period was stronger
where the gap is coming from
This analysis is extremely important because business
decisions happen through comparisons, not raw
numbers.
Use cases:
This quarter vs last quarter performance
Region A vs Region B sales
Paid vs organic customer quality
What it delivers:
A direct view of winners vs underperformers — making it
easier to allocate budget, teams, and strategy.
>
Poornachandra
Kongara
@poornachandra-kongara
Segmentation Analysis
(Who behaves differently?)
Segmentation analysis breaks one big dataset into
meaningful groups so you can understand different
behaviors.
Because in real companies:
Not all customers behave the same.
One segment may:
buy frequently
churn early
spend more
ask for discounts
generate more support tickets
So segmentation helps you understand the “why”
behind patterns.
Use cases:
High-value vs low-value customers
Churn rate by plan type
Retention by cohort
What it delivers:
Deep insight into customer types and behavior — helping
companies target the right users with the right strategy.
>
Poornachandra
Kongara
@poornachandra-kongara
Funnel Analysis (Where
do users drop off?)
Funnel analysis is a step-by-step breakdown
of how people move through a process.
It’s used to identify:
the biggest drop-off point
the step where people lose interest
where business loses conversions
A funnel makes it easy to see where the pipeline
“leaks” and where you need improvements.
Use cases:
Ecommerce: Add to cart → Checkout → Payment
SaaS onboarding funnel
Sales pipeline funnel
What it delivers:
A clear map of conversion bottlenecks — helping improve
revenue and user activation without needing more traffic.
>
Poornachandra
Kongara
@poornachandra-kongara
Cohort Analysis (How
groups behave over time?)
Cohort analysis studies groups of users who
started at the same time, and tracks their
behavior over weeks or months.
This is powerful because it helps answer:
Do new users behave better than old users?
Did retention improve after a feature launch?
Are customers sticking longer than before?
It reveals long-term patterns that daily trends often
hide.
Use cases:
retention tracking
churn pattern detection
product improvement impact
What it delivers:
A long-term retention and loyalty view — proving whether
product updates are actually improving user behavior.
>
Poornachandra
Kongara
@poornachandra-kongara
Correlation / Relationship
Analysis (What influences
what?)
Relationship analysis helps you explore
which factors might influence a KPI.
Here you try to discover:
what variables move together
which factors show strong relationship with outcomes
which drivers may impact churn, retention, revenue,
satisfaction
This analysis is often used to generate hypotheses for
deeper investigation.
Use cases:
Does discount increase sales?
Does app usage increase retention?
Does support tickets reduce customer satisfaction?
What it delivers:
A long-term retention and loyalty view — proving whether
product updates are actually improving user behavior.
Reminder:
Correlation shows patterns, not final proof.
>
Poornachandra
Kongara
@poornachandra-kongara
Root Cause Analysis
(Why did this happen?)
Root cause analysis is the most business-critical
analysis because it connects problems to actions.
When leadership asks:
“Why did this KPI drop?”
They don’t want charts — they want reasons.
This analysis involves:
drilling down by segments
identifying changes in behavior
isolating top reasons using breakdowns
validating what exactly triggered the change
Use cases:
Why did churn spike?
Why did revenue drop?
Why are returns increasing?
What it delivers:
The actual reason behind performance changes —
enabling teams to fix the real issue instead of guessing.
>
Poornachandra
Kongara
@poornachandra-kongara
Anomaly / Outlier Detection
(What looks unusual?)
fraud
system errors
reporting bugs
sudden behavior changes
unexpected spikes/drops
It is extremely important because many major
business risks show up as anomalies first.
Use cases:
payment failures
fraud detection
data quality issues
system outages
What it delivers:
Early warning signals — allowing businesses to detect
problems early before they cause large losses.
>
Poornachandra
Kongara
@poornachandra-kongara
Predictive Analysis (What
will happen next?)
Anomaly / Outlier Detection (What looks
unusual?)
Instead of reporting what happened, you
forecast what is likely to happen next using
historical patterns.
This supports planning and decision-making by
estimating:
future sales demand
churn probability
inventory needs
customer lifetime value trends
Use cases:
sales forecasting
churn prediction
demand planning
inventory planning
What it delivers:
Future-ready insights — helping teams plan better, reduce
risk, and take action before problems happen.
Poornachandra Kongara
Follow for more