Visualizations — Kab Kya Use Karein (Conditions Focus)

Power BI mein 30+ visual types available hain — lekin galat visual choose karna data story ko barbaad kar deta hai. Is part mein har visual type ke liye clear conditions sikhenge — "KAB use karo, KAB mat karo, KYUN use karo." Right visual = Right insight. Data Insights par complete guide.

📑 Is Part Mein Aap Kya Sikhenge:

Har visual type ke liye — kab use karein, kab avoid karein, best practices aur configuration tips:

• Card, KPI, Multi-row Card: Single number display — executive dashboards ki foundation
• Table vs Matrix: Raw data vs cross-tabulation — kab kya fit hota hai
• Bar vs Column Charts: Category comparison — horizontal vs vertical kab use karein
• Line vs Area Charts: Time series trends — continuous data ke liye
• Pie vs Donut vs Treemap: Part-to-whole analysis — limitations aur alternatives
• Scatter vs Bubble: Correlation & distribution — analytical visuals
• Gauge vs KPI Visual: Progress & target tracking — performance indicators
• Map Visuals: Geographic data — Filled Map, ArcGIS, Shape Map

📋 Visual Selection Framework — Pehle Yeh Pucho:

1. Kya dikhana hai? — Single number? Comparison? Trend? Distribution? Proportion? Location?
2. Kitne data points hain? — 1 number? 5 categories? 100 rows? Thousands?
3. Audience kaun hai? — Executives (simple KPIs)? Analysts (detailed tables)? Public (clean charts)?
4. Action kya lena hai? — Monitor progress? Compare regions? Find trends? Drill into details?

🔍 Definition: Card visual displays a single large number — the most prominent value on a dashboard (Total Sales, Total Orders, Revenue). KPI Visual shows a value with a target and trend indicator — progress toward a goal. Multi-row Card shows multiple KPIs in a single compact visual — multiple measures in one box. These are the first things executives look at on any dashboard — the "headline numbers."

🎯 Samjho Hinglish Mein: Card ek scoreboard hai — sirf ek bada number dikhata hai. Jaise stadium mein score dikhta hai "India: 285." KPI score ke saath target bhi dikhata hai — "285 / 300 Target = 95% achieved." Multi-row Card ek mini dashboard hai — ek box mein 4-5 numbers dikha deta hai: Total Sales, Orders, Avg Value, Customers — sab ek jagah. Executives ko pehle yeh headline numbers chahiye — details baad mein.

💡 Kab Use Karein — Conditions:
• Card ✅ Use When: Ek single important metric highlight karna hai (Total Revenue, Customer Count). Dashboard ke top mein headline numbers ke liye. Slicer ke saath dynamic total dikhana ho.
• Card ❌ Avoid When: Multiple metrics ek saath dikhane hain (Multi-row Card use karo). Trend ya comparison dikhana hai (chart use karo).
• KPI ✅ Use When: Target vs Actual comparison chahiye. Progress tracking chahiye (sales target achievement). Trend direction (up/down arrow) dikhana ho.
• KPI ❌ Avoid When: Target value defined nahi hai. Multiple KPIs chahiye (multi-row ya separate cards better).
• Multi-row Card ✅ Use When: 3-6 metrics ek compact space mein dikhane hain. Dashboard header section mein summary strip chahiye.

📊 Card Types — Comparison:

💻 Configuration — What to Drag Where:

// ═══════════════════════════════════════════ 
// CARD VISUAL
// ═══════════════════════════════════════════ 
// Fields well: "Fields" → drag [Total Sales] measure 
// Format: Display units = Thousands/Millions/Auto
// Format: Category label = ON (shows measure name below number) 
// Tip: Create separate cards for Total Sales, Order Count,
// Unique Customers — line them up horizontally at top
// ═══════════════════════════════════════════
// KPI VISUAL
// ═══════════════════════════════════════════
// Indicator: [Total Sales] (the actual value)
// Target: [Sales Target] (the goal value)
// Trend Axis: Calendar[MonthName] (for sparkline)
// 
// Target measure example:
Sales Target = 500000
// Or dynamic target from a Targets table:
Sales Target = SUM(Targets[TargetAmount])

// ═══════════════════════════════════════════
// MULTI-ROW CARD
// ═══════════════════════════════════════════
// Fields: drag multiple measures:
// → [Total Sales]
// → [Order Count]
// → [Avg Order Value]
// → [Unique Customers]
// Format: Word wrap = ON, Text size = adjust

⚠️ Common Mistakes:

• Mistake: Card mein column drag karna measure ki jagah — unexpected aggregation dikhta hai. Fix: Hamesha measures drag karo cards mein. Column drag karne par Power BI auto-sum ya count kar deta hai — galat result.
• Mistake: Display units "Auto" rakhna — ₹1645000 ugly dikhta hai. Fix: Display units "Thousands" ya "Lakhs" set karo → "₹16.5L" dikhega. Professional dashboards mein abbreviated numbers use hote hain.
• Mistake: Bahut saare individual cards banana — dashboard cluttered ho jaata hai. Fix: 4-6 se zyada top-level KPIs mat rakho. Baaki Multi-row Card ya detail page mein rakho.

💬 Interview Questions:

🔍 Definition: Table visual displays data in a flat, row-by-row format — like an Excel spreadsheet. Each row is one record. Matrix visual displays data in a cross-tabulated pivot format — rows AND columns both have categories, with values at intersections. Matrix supports hierarchies, expand/collapse, subtotals, and stepped layout. Table is for detailed record viewing, Matrix is for summarized cross-analysis.

🎯 Samjho Hinglish Mein: Table = register — har order ek row mein, flat list. "OrderID 1001, Laptop, North, ₹1,10,000." Matrix = pivot table — rows mein Region, columns mein Category, beech mein Sales. "North × Electronics = ₹3,10,000." Table raw details ke liye, Matrix summarized analysis ke liye. Executives ko Matrix chahiye (summary), analysts ko Table chahiye (details).

💡 Decision Conditions:
• Table ✅ When: Raw detail records dikhane hain (order list, transaction log). Drill-through detail page mein. Export ke liye data preview. Column count > row groupings.
• Table ❌ Avoid: Summarized cross-analysis chahiye. Hierarchies drill karne hain. Subtotals chahiye category wise.
• Matrix ✅ When: Pivot-style analysis chahiye (Region × Product = Sales). Hierarchies expand/collapse chahiye (Year → Quarter → Month). Row + Column subtotals chahiye. Conditional formatting apply karni hai cells par.
• Matrix ❌ Avoid: Raw individual records dikhane hain. Simple flat list chahiye bina grouping ke.

📊 Table vs Matrix — Comparison:

💻 Configuration Tips:

// TABLE VISUAL:
// Columns: OrderID, OrderDate, Customer, Product, Amount
//
Values: drag columns you want to display
// Format: Totals →
ON/OFF, Alternating rows → better readability
// Conditional Formatting: Amount column → Background color scale
// MATRIX VISUAL:
// Rows: Region (or hierarchy: Year → Quarter → Month)
// Columns: Category (Electronics, Accessories, Furniture)
//
Values: [Total Sales]
//
// Result layout:
// Electronics Accessories Furniture TOTAL
// North ₹3,10,000 ₹85,000 ₹1,25,000 ₹5,20,000
// South ₹2,40,000 ₹60,000 ₹2,00,000 ₹5,00,000
// East ₹1,50,000 ₹40,000 ₹1,50,000 ₹3,40,000
// West ₹1,30,000 ₹30,000 ₹1,25,000 ₹2,85,000
// TOTAL ₹8,30,000 ₹2,15,000 ₹6,00,000 ₹16,45,000

⚠️ Common Mistakes:

• Mistake: Summary page par Table use karna — bahut rows dikhai deti hain, cluttered. Fix: Summary pages ke liye Matrix use karo with grouping. Table sirf detail/drill-through pages ke liye.
• Mistake: Matrix mein bahut saare columns drag karna — unreadable ban jaata hai. Fix: Matrix mein 3-5 column groups se zyada mat rakho. Zyada categories hain toh Rows mein rakho.
• Mistake: Conditional formatting nahi lagana Matrix mein. Fix: Data bars ya color scales lagao — patterns instantly visible ho jaate hain. Highest value dark green, lowest light red.

💬 Interview Questions:

🔍 Definition: Column Chart (vertical bars) and Bar Chart (horizontal bars) are used for comparing values across categories. Column charts have categories on X-axis and values on Y-axis. Bar charts have categories on Y-axis and values on X-axis. Both support Clustered (side-by-side), Stacked (on top of each other), and 100% Stacked (percentage distribution) variants.

🎯 Samjho Hinglish Mein: Column Chart = vertical bars — "North, South, East, West ki sales compare karo." Bar Chart = horizontal bars — same comparison lekin sideways. Kab kya use karo? Agar category names lambe hain ("Customer Service Department") — horizontal Bar Chart better hai kyunki labels clearly padhne aate hain. Agar categories kam hain (4 regions) — Column Chart clean dikhta hai. Agar time axis hai — Column Chart preferred hai (time left-to-right natural flow).

💡 Decision Conditions:
• Column Chart ✅ When: Categories kam hain (3-8). Category names chhote hain. Time-based comparison (months, quarters). Natural left-to-right reading flow chahiye.
• Bar Chart ✅ When: Categories zyada hain (8-15+). Category names lambe hain. Ranking dikhana hai (longest bar = highest). Top/Bottom N analysis.
• Clustered ✅ When: 2-3 groups compare karne hain side-by-side (2023 vs 2024 per region).
• Stacked ✅ When: Total + breakdown dikhana hai (each region's total with category breakdown).
• 100% Stacked ✅ When: Proportions compare karne hain (each region's category percentage split).

📊 Bar/Column Variants — Quick Guide:

⚠️ Common Mistakes:

• Mistake: 15 categories Column Chart mein — labels overlap ho jaate hain, unreadable. Fix: 8+ categories ke liye Bar Chart (horizontal) use karo — labels clearly dikhte hain.
• Mistake: Clustered mein 5 legend items — bars itne thin ho jaate hain ki compare nahi ho paata. Fix: Clustered mein max 2-3 legend items rakho. Zyada hain toh Stacked use karo.
• Mistake: Bar chart mein sort nahi karna — random order confusing. Fix: Hamesha sort karo — ascending ya descending by value. Ranking clear dikhta hai.
• Mistake: Stacked chart mein individual segment comparison karna — middle segments ka base different hai. Fix: Sirf bottom segment accurately compare ho paata hai. Individual comparison ke liye Clustered use karo.

💬 Interview Questions:

🔍 Definition: Line Chart connects data points with lines to show trends over time. Area Chart is a line chart with the area below the line filled with color — emphasizes volume/magnitude. Both are perfect for continuous time-series data. Line charts are better for comparing multiple series (2023 vs 2024 lines). Area charts are better for showing cumulative volume or part-to-whole over time (stacked area).

🎯 Samjho Hinglish Mein: Line Chart = trend line — "Sales monthly kaise badh rahi hain ya ghatt rahi hain?" Ek line 2024 ki, ek 2023 ki — clearly compare ho jaata hai. Area Chart = same line lekin neeche color fill — volume ka ehsaas hota hai. Stacked Area mein multiple categories ka total contribution dikhta hai over time. Line Chart zyada common hai dashboards mein. Area Chart tab use karo jab volume/magnitude emphasize karni ho.

💡 Decision Conditions:
• Line Chart ✅ When: Time-based trends dikhane hain. Multiple series compare karne hain (2023 vs 2024). Running Total / YTD line. Precise values compare karne hain.
• Area Chart ✅ When: Volume/magnitude emphasize karni hai. Stacked Area — category contribution over time. Cumulative data dikhana hai.
• Both ❌ Avoid When: Data categorical hai (non-time) — Bar/Column better. Data points bahut kam hain (3-4) — Bar Chart better. Discrete comparison chahiye — Bar Chart better.

📊 Line vs Area — When to Use:

⚠️ Common Mistakes:

• Mistake: Line Chart mein 8+ lines — spaghetti mess ban jaata hai. Fix: Max 3-4 lines ek chart mein. Zyada series hain toh Small Multiples ya separate charts use karo.
• Mistake: Area Chart mein multiple non-stacked areas — overlap se kuch dikhta nahi. Fix: Multiple areas ke liye Stacked Area use karo. Ya Line Chart switch karo.
• Mistake: Time axis mein gaps hona (missing months). Fix: X-axis → Type = "Continuous" set karo (Categorical nahi). Date Table ki continuous dates use karo.

💬 Interview Questions:

🔍 Definition: Pie Chart shows parts of a whole as slices of a circle — each slice's angle represents its proportion. Donut Chart is a Pie Chart with a hole in the center — the hole can display a total or KPI value. Treemap shows parts of a whole as nested rectangles — larger rectangles represent larger values, and it supports hierarchies. All three answer: "What percentage of the total does each category represent?"

🎯 Samjho Hinglish Mein: Pie = pizza slices — "Electronics ka slice kitna bada hai total mein?" Donut = Pie with a hole (beech mein total likh sakte ho). Treemap = rectangles — bade rectangle = badi value, chhota rectangle = chhoti value. Important: Pie Chart ki bahut criticism hoti hai data visualization community mein — 5+ slices mein compare karna mushkil ho jaata hai. Treemap better alternative hai 6+ categories ke liye.

💡 Decision Conditions:
• Pie/Donut ✅ When: Categories 2-5 hain MAX. Proportions clearly different hain (60% vs 30% vs 10%). Executive summary mein simple share dikhana hai.
• Pie/Donut ❌ Avoid When: Categories 6+ hain — slices too thin, unreadable. Values bahut similar hain (25% vs 24% vs 23%) — visually indistinguishable. Precise comparison chahiye — Bar Chart better.
• Treemap ✅ When: Categories 5-20+ hain. Hierarchical data hai (Category → Sub-category). Space-efficient display chahiye. Quickly largest/smallest spot karna hai.
• Best Alternative: 100% Stacked Bar/Column — sab se reliable part-to-whole visual. Proportions clearly compare ho paate hain.

📊 Pie vs Donut vs Treemap — Comparison:

⚠️ Common Mistakes:

• Mistake: 10 slices ka Pie Chart — unreadable, colors indistinguishable. Fix: Max 5 slices. Zyada hain toh top 4 + "Others" category banao ya Treemap use karo.
• Mistake: Pie Chart mein data labels nahi lagana — kaunsa slice kitna hai pata nahi. Fix: Hamesha percentage labels ON karo. Category name + % dono dikhao.
• Mistake: Two Pie Charts compare karna (side by side) — ineffective. Fix: Two periods compare karne ke liye 100% Stacked Bar use karo — comparison clear hota hai.

💬 Interview Questions:

🔍 Definition: Scatter Chart plots data points on X-Y axes to show relationships/correlation between two numeric variables. Each dot represents one entity (product, customer, region). Bubble Chart is a Scatter Chart where the size of each dot (bubble) represents a third variable — adding a third dimension of data. Both are analytical visuals used to find patterns, clusters, outliers, and correlations.

🎯 Samjho Hinglish Mein: Scatter = "Kya zyada ad spend wale products ki zyada sales hain?" X-axis par Ad Spend, Y-axis par Sales — agar dots bottom-left se top-right ki taraf jaa rahe hain toh positive correlation hai (zyada spend = zyada sales). Bubble = same scatter lekin dot ka size profit represent karta hai — bada bubble = zyada profit. Isse ek chart mein 3 metrics dikh jaate hain. Yeh analytical tool hai — executives se zyada analysts use karte hain.

💡 Decision Conditions:
• Scatter ✅ When: Do numeric variables ka correlation dekhna hai. Outliers identify karne hain. Clusters find karne hain (customer segments). Data points 10+ hain.
• Bubble ✅ When: Teen numeric variables ek chart mein dikhane hain. Size = third metric (revenue, profit, quantity).
• Both ❌ Avoid When: Categories compare karne hain (Bar Chart use karo). Time trend dikhana hai (Line Chart use karo). Data points 5 se kam hain (Table better). Non-technical audience — confusing lag sakta hai.

💻 Configuration:

// SCATTER CHART: 
// X Axis: [Avg Order Value] (measure) 
// Y Axis: [Order Count] (measure) 
// Details: Products[ProductName] (each dot = one product) 
// Legend: Products[Category] (color by category) 
// 
// Interpretation: 
// Top-right quadrant: High AOV + High Orders = STARS ⭐ 
// Top-left: Low AOV + High Orders = Volume drivers 
// Bottom-right: High AOV + Low Orders = Premium niche 
// Bottom-left: Low AOV + Low Orders = UNDERPERFORMERS ❌

// BUBBLE CHART (add Size):

// X Axis: [Avg Order Value]

// Y Axis: [Order Count]

// Size: [Total Sales] (bubble size = revenue)

// Details: Products[ProductName]

// Legend: Products[Category]

//

// Now 3 metrics in ONE chart:

// Position (X,Y) + Size = AOV, Orders, Revenue

⚠️ Common Mistakes:

• Mistake: Scatter mein Details field bhoolna — ek bada dot aata hai sab data ka. Fix: Details field mein woh column drag karo jisse har entity ek dot bane (ProductName, CustomerName).
• Mistake: Bahut zyada dots (1000+) — overplotting, sab overlap. Fix: Dot transparency badhao, ya aggregated data use karo (product-level instead of order-level).
• Mistake: Bubble sizes bahut similar — koi pattern nahi dikhta. Fix: Size metric mein enough variance chahiye. Ya Scatter (without size) use karo.

💬 Interview Questions:

🔍 Definition: Gauge visual shows a value on a semicircular scale with a minimum, maximum, and optional target — like a car speedometer. It shows how close a value is to a target. KPI Visual shows a value with target comparison and a trend sparkline — it answers "Are we on track?" and "What's the trend direction?" Both are used for target/goal tracking but in different ways.

🎯 Samjho Hinglish Mein: Gauge = car ka speedometer — needle current value par hai, green zone target hai. Ek nazar mein pata chal jaata hai "target ke kitne paas hain." KPI Visual = scorecard — number dikhata hai, uske neeche target, aur ek mini trend line. Gauge visual impact ke liye — "look, we're almost there!" KPI precision ke liye — exact values + trend direction. Dono target tracking ke liye hain lekin presentation style different hai.

💡 Decision Conditions:
• Gauge ✅ When: Single metric ka target achievement dikhana hai visually. Min/Max range defined hai. Dashboard mein visual impact chahiye. "Speedometer" feel chahiye.
• Gauge ❌ Avoid When: Multiple metrics track karne hain (space inefficient). Exact numbers zyada important hain visual se. Trend direction chahiye (Gauge mein trend nahi dikhta).
• KPI Visual ✅ When: Value + Target + Trend teeno chahiye ek visual mein. Compact space mein goal tracking. Trend direction (improving/declining) important hai.
• Alternative: New Card visual with reference values — modern alternative to both.

💻 Configuration:

// GAUGE VISUAL:

// Value: [Total Sales] (needle position)

// Minimum: 0

// Maximum: [Sales Target] × 1.2 (120% of target for headroom) 

// Target: [Sales Target] (target line on gauge) 

// 

// Format: Conditional colors 

// 0-70% of target: Red

// 70-100%: Yellow

// 100%+: Green

// KPI VISUAL:

// Indicator: [Total Sales] (actual value)

// Target: [Sales Target] (comparison value)

// Trend Axis: Calendar[Date] or Calendar[MonthName]

//

// Shows: value, % of target, green/red indicator, mini sparkline

📊 Gauge vs KPI — Comparison:

⚠️ Common Mistakes:

• Mistake: Multiple Gauges side-by-side — bahut space waste hota hai. Fix: 1-2 gauges max ek dashboard par. Zyada KPIs ke liye Cards ya KPI visuals use karo.
• Mistake: Gauge ka Maximum set nahi karna — auto-max bahut bada ho jaata hai, needle chhoti dikhai deti hai. Fix: Maximum = Target × 1.2 set karo. Needle ka proportion meaningful dikhna chahiye.
• Mistake: Gauge mein target line nahi dena — just a dial without reference point. Fix: Target field zaroori hai — bina target ke gauge meaningless hai, sirf number dikha raha hai.

💬 Interview Questions:

🔍 Definition: Map visuals display data geographically on a map. Power BI offers several types: Map (Bubble Map) — plots bubbles on locations, bubble size = value. Filled Map (Choropleth) — colors entire regions/countries based on values. ArcGIS Map — advanced mapping with multiple layers, heat maps, clustering. Shape Map — custom shapes (states, custom boundaries). Maps are used when location is a key dimension of analysis.

🎯 Samjho Hinglish Mein: "India ke kaunse state mein sabse zyada sales hain?" — yeh answer Map visual deta hai. Bubble Map mein har city par ek bubble aata hai — bada bubble = zyada sales. Filled Map mein states ko color karta hai — dark green = zyada sales, light = kam. ArcGIS Map advanced hai — heat maps, reference layers, custom analysis. Jab location data hai (City, State, Country, Lat/Long) tab Map visuals use karo.

💡 Decision Conditions:
• Map (Bubble) ✅ When: Point locations dikhane hain (cities, stores, offices). Bubble size se value represent karni hai. Specific coordinates (Lat/Long) available hain.
• Filled Map ✅ When: Regions/states/countries color-code karne hain by value. Administrative boundaries compare karne hain. "Which region performs best?" answer karna hai.
• ArcGIS ✅ When: Advanced geo-analysis chahiye — heat maps, drive-time areas, clustering, multiple layers. Premium geographic insights.
• All Maps ❌ Avoid When: Location data nahi hai ya irrelevant hai. Precise value comparison chahiye (Bar Chart better — map pe bubble sizes compare karna hard hai). Space limited hai (maps space lete hain).

📊 Map Types — Comparison:

💻 Configuration Tips:

// MAP (BUBBLE) VISUAL: // Location: Customers[City] (or Lat/Long columns) // Size: [Total Sales] (bubble size = sales value) // Legend: Products[Category] (color by category) // Tooltips: [
Order Count], [Avg
Order Value] // // IMPORTANT:
Set Data Category
on location columns! // Data View → Select City column → Column Tools → // Data Category → "City" // Without this,
Bing Maps may plot wrong locations!
// FILLED MAP:
// Location: Customers[State] (or Country)
// Legend / Color Saturation: [Total Sales]
// Tooltips: [
Order Count], [YoY Growth %]
// Data Category: "State or Province" (must set!)

⚠️ Common Mistakes:

• Mistake: Data Category set nahi karna — map galat locations dikhata hai. Fix: Data View mein har geographic column ki Data Category set karo (City, State, Country, Latitude, Longitude).
• Mistake: Map par precise value comparison karna chahna — bubbles ka size accurately compare nahi hota. Fix: Map geographic distribution ke liye best hai. Precise comparison ke liye Bar Chart use karo. Map + Bar Chart combo dashboard mein rakho.
• Mistake: Map enable nahi hai — blank visual dikhta hai. Fix: File → Options → Security → Map and filled map visuals → Enable. ArcGIS ke liye separate enable karo.
• Mistake: Bahut saare data points (1000+ cities) — map slow aur cluttered. Fix: Data aggregate karo (state-level), ya ArcGIS clustering use karo. Map visuals performance heavy hain — wisely use karo.

💬 Interview Questions:

"Kya dikhana hai?" ke basis par right visual choose karo:

Next: Power BI Masterclass — Part 7 (Final)

Final part mein hum cover karenge: Advanced Features & Interactivity — Slicers (Types & Best Use Cases), Drill-through vs Drill-down, Bookmarks & Buttons (Navigation), Custom Tooltips, Conditional Formatting in Visuals, Hierarchies, Sync Slicers Across Pages, aur Row Level Security (RLS) Theory. Yeh sab features dashboard ko interactive aur professional banate hain. Previously completed: MySQL, Pandas, NumPy, Data Cleaning, Matplotlib, Seaborn, Plotly, Excel Masterclass — sab Data Insights par available hai.

Happy Learning & Keep Analyzing! 🚀