📚 Prerequisites & Resources — Data Analyst Job Simulation

Read this before starting the clock. The simulation assumes zero work experience, but it does assume you can operate a spreadsheet. If anything in the checklists below is new to you, spend 2–4 hours with the resources first — the timer is unforgiving, the prep is free.

Who this is for

Freshers and career-switchers targeting Data Analyst / Business Analyst / MIS Analyst roles. If you can do both checklists below, you're ready. If you can't yet — that's exactly what this page fixes.

⏱️ The rules at a glance

Rule	What it means for you
The clock	Starts when you start a ticket, pauses while your work is in review, and resumes on retry. Your total active time is what ranks you on the weekly leaderboard.
First submission — 12 hours	Make at least one submission within 12 hours of starting, or the simulation deactivates.
Everything — 48 hours	Submit all tickets within 48 hours of starting. Plan it like a real take-home: one sitting per ticket.
Free submissions	2 per ticket. After that: one instant retry for ₹49, or wait 6 hours for a free slot.
Your dataset	Generated uniquely for you and stays the same for your whole run — a retry means fixing your existing file, not starting over. Copied answers from anyone else's file will fail automatically.
Deactivated?	Restart free after 7 days, or instantly for ₹99. A restart issues a fresh dataset and resets progress.
Results	Within 24 hours, usually much sooner — you'll get an email plus in-app status the moment grading finishes.

🧮 How you're graded (pass: 60/100 on DA-101, 70/100 on DA-102)

DA-101: 70% automated check of your cleaned file against the known error list (every injected error, fix by fix — and modifying rows that were already clean loses marks), 30% AI review of your data-quality note (≥3 issue types, counts, drop justification, clarity).
DA-102: 60% automated numeric answers computed from your dataset (tolerances are printed in the ticket), 40% AI review of your memo — where decision quality (did you look beyond the single metric you were given?) is worth as much as being right.
Feedback always tells you the score math and exactly which operations lost marks, so a retry is never a guess.

✅ Skill checklist — Ticket DA-101 (Data Cleaning)

You should be able to do each of these in Excel or Google Sheets (Python equivalents shown for those going that route):

Skill	Excel / Sheets	Python (pandas)
Sort & filter a table	Data → Filter	`df.sort_values()`, boolean masks
Remove exact duplicate rows	Data → Remove Duplicates	`df.drop_duplicates(keep="first")`
Trim stray spaces	`TRIM()`	`.str.strip()`
Fix casing against an official list	`=XLOOKUP(LOWER(TRIM(A2)), LOWER(list), list)` — or plain Find & Replace per broken value	build a dict from the official names: `fix = {n.lower(): n for n in official}` then `.str.strip().str.lower().map(fix)`
Find & replace specific values	Ctrl+H, or `SUBSTITUTE()`	`.replace({...})`
Standardize dates	helper formulas — see the date recipe below	`pd.to_datetime(col, dayfirst=True, format="mixed")` (pandas ≥ 2.0), then `.dt.strftime("%Y-%m-%d")`
Convert text to numbers	`VALUE()` + `SUBSTITUTE()` to strip ₹, commas, "INR"	`.str.replace(r"[₹,]
Find blank cells	Filter by (Blanks), `ISBLANK()`	`df.isna()`
Fill values via a lookup table	`VLOOKUP()` / `INDEX`+`MATCH` (city → region)	`.map(city_to_region)`
Delete rows matching a condition	Filter → select rows → delete	`df = df[df["quantity"] > 0]`

⚠️ The #1 mark-loser: blanket PROPER() / .str.title(). Several official product names use deliberate casing (43-inch Smart TV, Power Bank 20000mAh, Sunscreen SPF50…). A formula that title-cases every row silently corrupts ~a quarter of the file — and the grader counts each one as a wrongly modified clean row. Fix only the broken rows, by matching them to the official spellings printed in the ticket.

📅 The date recipe (Excel/Sheets). The file mixes three broken formats, and DATEVALUE() behaves differently per locale — so parse them yourself with helper columns, then format everything =TEXT(cell,"YYYY-MM-DD"):

08/03/2026 (day first) → =DATE(RIGHT(A2,4), MID(A2,4,2), LEFT(A2,2))
08-03-26 (day first, 2-digit year) → =DATE(2000+RIGHT(A2,2), MID(A2,4,2), LEFT(A2,2))
Mar 8, 2026 → DATEVALUE() handles this one fine in most locales
Sanity check when done: every date must fall in Jan–Jun 2026. A date like 2026-08-03 means you swapped day and month.

📤 Deliverable format (the auto-grader is literal): same 11 columns, same names, same order, no index column (pandas: to_csv(index=False)), quantity/total_revenue as plain numbers (1751, not 1751.0 or ₹1,751), dates as YYYY-MM-DD text.

🔎 Self-check before submitting: every surviving row should satisfy quantity × unit_price = total_revenue. One failing row = a parsing error you haven't found yet.

Self-test: given a column with " MUMBAI ", bombay, and Mumbai, can you make all three read Mumbai — without touching a fourth cell that's already correct? If yes in under 2 minutes, you're ready for DA-101.

✅ Skill checklist — Ticket DA-102 (Analysis & Recommendation)

Work from your cleaned DA-101 file — a wrong answer here usually traces back to a cleaning mistake there (wrong drops shift every total).

Skill	Excel / Sheets	Python (pandas)
Summarize revenue by a category	Pivot Table (rows: region, values: SUM of revenue) or `SUMIFS()`	`df.groupby("region")["total_revenue"].sum()`
Group by month from a date	Pivot: Group dates by month, or a helper column `=TEXT(date,"YYYY-MM")`	`df["order_date"].str[:7]` or `.dt.to_period("M")`
% change between two numbers	`=(new-old)/old`, format as %	`.pct_change()`
% share of total	`=part/SUM(total)`	`x / x.sum()`
Average Order Value (AOV)	`=SUMIFS(revenue,channel,X)/COUNTIFS(channel,X)`	`groupby()["revenue"].sum() / groupby().size()`
Compare a metric across months for one segment	Pivot with two dimensions (month × channel)	`pivot_table(index="month", columns="channel", values="total_revenue", aggfunc="sum")`
Write a short business memo	Recommendation first → 2+ metrics as evidence → one next step. ~150 words, readable in 40 seconds.	—

Answer formats matter: percentages to 1 decimal place, revenue as a plain number, direction as exactly one of Growing / Declining / Mixed. The tolerances (±1% on revenue, ±2 points on MoM %, ±1 point on share %) are printed in the ticket — they forgive rounding, not method errors.

Concepts to know (60-second definitions below): MoM growth, AOV, revenue share, trend vs. snapshot.

📖 Mini glossary

MoM (Month-on-Month) growth — % change vs the previous month: (this − last) ÷ last × 100. The board's favorite momentum metric.
AOV (Average Order Value) — total revenue ÷ number of orders. Tells you how much a typical order is worth, independent of how many orders there are.
Revenue share — one segment's revenue as a % of the whole. A snapshot.
Trend vs. snapshot — a snapshot says where something is; a trend says where it's going. Decisions made on snapshots alone are how companies kill their fastest-growing channel. (That's a hint.)
Exact duplicate — a row identical in every column, usually a system export error.
Canonical value — the one official spelling of a thing (Bengaluru, not BLR; 43-inch Smart TV, not 43-Inch Smart Tv).

🚫 Top 5 ways people lose marks (all avoidable)

Blanket title-casing product names — corrupts deliberately-cased names at scale. Fix only broken rows, against the official lists.
Filling what should be dropped — blank quantity/revenue rows and zero/negative quantities must be deleted, per the finance rule in the ticket. Blank region is the only fill.
Day/month swaps — 08/03/2026 is 8 March, not 3 August. Every date lands in Jan–Jun 2026; anything outside that range is your parser lying to you.
A 12th column — pandas' index column (to_csv(index=False) forgets) or an Excel helper column left in. The submit check rejects it before grading.
Modifying clean rows — every unnecessary "fix" costs marks. When in doubt, leave it alone.

🔗 Free resources

Spreadsheets

Microsoft Excel help & function reference — https://support.microsoft.com/en-us/excel
Google Sheets Help Center — https://support.google.com/docs
W3Schools Excel tutorial (fast, hands-on) — https://www.w3schools.com/excel/
Search on YouTube: "Excel pivot table tutorial", "TRIM SUBSTITUTE Excel", "remove duplicates Excel", "XLOOKUP tutorial" — any top freeCodeCamp / Leila Gharani result covers what you need.

Python route (optional)

Kaggle Learn — Pandas (free micro-course, ~4 hrs) — https://www.kaggle.com/learn/pandas
freeCodeCamp YouTube channel — https://www.youtube.com/@freecodecamp

Structured deep-dive (optional, not required to pass)

Google Data Analytics Professional Certificate (audit for free) — https://www.coursera.org/professional-certificates/google-data-analytics

From OneRoadmap (replace with live links at launch)

📄 Data Analyst Fresher Guide (PDF) — (#)
🎓 Data Analyst "Big 4 Ready" Certification — the natural next step after clearing this simulation — (#)
📄 Excel Interview Questions Pack — (#)

🏋️ 30-minute warm-up drill (recommended)

Make a 20-row fake sales table with deliberate mess: 2 duplicate rows, 3 date formats, "₹1,200" as text, two blank cells, " delhi ", and one product name with intentional casing (e.g. Power Bank 20000mAh typed as power bank 20000mah).
Clean it using only the checklist above — no googling during the drill. Fix the product name back to its exact official spelling without title-casing the whole column.
Build one pivot: revenue by category. Compute one % change and one % share.

Under 30 minutes and correct? Start the simulation. Over 30 minutes? Spend an evening with the resources first — the leaderboard clock counts everything, including panic-googling XLOOKUP.

⚙️ Practical setup

Works with: Excel (2016+), Google Sheets (free), LibreOffice Calc, or Python. We grade the output, never the tool.
The dataset is a ~5,000-row CSV (~500 KB) — any laptop handles it. On mobile? Google Sheets works, but a laptop is strongly recommended.
Plan your sitting before you press Start: download → clean → self-check → submit in one go beats coming back cold. The clock pauses during review, so the gap between tickets costs you nothing.