Available · India · UAE · Singapore
Levi Strauss & Co. 2 yr
BITS Pilani Chem Engg · Finance
Harvard PAIR 1 of 240

Devansh
Pandey

Data  ·  Strategy  ·  Product  ·  Analytics

I build systems that improve commercial decisions at scale. At Levi Strauss & Co., I help drive inventory decisions across 400+ retail stores, balancing full-price sell-through, markdowns, and replenishment across India.

◈
How I think: Most product problems have a measurement problem inside them. The work is finding the right number to track before anything gets built.
2yr
Levi Strauss & Co.
1 of 240
Harvard PAIR Delegates
BITS Pilani
Chem Engg · Finance Minor
Scroll
Current profile
Role
Business Analyst
Company
Levi Strauss & Co. · Bengaluru
Scope
400+ stores · NCS Allocation
Buy plan
₹350Cr · India merchandise
Education
BITS Pilani · Chem Engg + Finance
Recognition
Harvard PAIR · 1 of 240 delegates
Target
PM · Strategy · Analytics
About

Building systems that help
businesses decide better.

2 years at Levi Strauss & Co., building the inventory intelligence that determines what 400+ retail stores across India carry, when they restock, and how much sells at full price versus markdown. I own the allocation logic and contribute to a ₹350 Cr India-wide merchandise buy plan, working across commercial, finance, and planning teams.

The core of the work: turning demand signals, seasonality patterns, and sell-through trends into allocation decisions that reduce waste and protect margin. I have built forecasting-backed workflows, a Tableau dashboard used by cross-functional business teams for merchandise visibility, and data pipelines that cut the time from data to decision.

Selected as 1 of 240 delegates at Harvard PAIR Tokyo 2025. NextLeap PM Fellow. At Levi's I work cross-functionally with commercial, finance, and planning teams, which means I have learned to translate between business strategy and data systems, a skill I find more useful than any single technical tool.

Product Strategy Demand Analytics User Research SQL · Python Power BI NextLeap PM Harvard PAIR BITS Pilani
Timeline

Levi Strauss & Co.

Business Analyst · Bengaluru
Jan 2024 – Present

Harvard PAIR Tokyo 2025

Selected Delegate · AI & Policy
2025

NextLeap PM Fellowship

Product Management Certification
2024

ANS Commerce

Business Analyst Intern · Category RCA
2023

BITS Pilani, Hyderabad

B.E. Chemical Engineering · Finance Minor
2020 – 2024

What I actually
do at Levi's.

Two tools, one reporting layer, and a lot of decisions about what goes where. This is the work behind 400+ chain stores and 1,200+ wholesale partners across India.

◈
Buy Plan Tool — built with manager input and business team collaboration Partners book inventory 6 months ahead. The problem: buyer decisions were not backed by a unified data picture. I built a tool where buyers input what they want to commit, and it populates that across 1,200+ wholesale stores, incorporating business goals, MER, trend analysis, brand story, regional preferences, size curves, and partner demand signals automatically. Measured a 15% improvement in sales rate comparing periods before and after adoption.
◈
Allocation Tool — 400+ chain stores After buying is locked in SAP, inventory sits in the warehouse. I decide what goes from warehouse to which store. The problem before: allocation reflected what was ordered 6 months ago. My tool incorporates what is actually selling right now — seasonality, regional trends, size curves, current store-level performance, and external signals. Allocation becomes a current-data decision, not a historical one.
◈
Reporting and analytics Metrics business leaders need, operational visibility, and ad hoc analysis. The connective tissue between strategy and execution.
Artifacts

How the tools
actually work.

System diagrams for the two tools I built at Levi Strauss — the buy plan tool and the allocation tool.

Buy Plan Tool — 1,200+ Wholesale Stores
The problem: buyer decisions not backed by unified dataWhat to buy is enteredQuantity committedTOOL INCORPORATES ALL OF THESEPast salesHistorical dataBusiness goalsMER · targetsTrend analysisBrand story directionPartner demandWhat stores wantRegional trendsState & city levelSize curvesBy geographyBrand storyWhere we want to goCustomer viewSegment trendsBuy Plan ToolPopulates store-level plan automaticallyStore-level Buy Plan1,200+ wholesale stores · data-backed+15% sales rate improvementMeasured before vs after tool adoptionBuy Plan Tool · Levi Strauss India
Allocation Tool — 400+ Chain Stores
The problem: allocation based on 6-month-old booking dataBuying locked in SAPPartner orders · 6 months priorWarehouse InventoryWhat was manufacturedCURRENT SIGNALS I INTRODUCEWhat is sellingright nowSeasonalitySize curvesRegionaltrendsExternaleventsAllocation DecisionCurrent-data plan for what goes whereStore-wise Allocation Plan400+ chain stores · with current-data tweaksFaster decisions backed by current dataNot what was planned 6 months agoAllocation Tool · Levi Strauss India
Case Studies

How I think
through problems.

One executed project and two analytical teardowns. The Levi Strauss case is work I actually shipped. Razorpay and PhonePe are structured product thinking exercises.

Internal Project · Retail Operations
Levi Strauss & Co.
Two tools, two problems: making current data drive six-month-old inventory decisions
· 14 min read  ·  ✓ Executed project
The obstacle: Buying is locked six months ahead. By the time goods arrive, the market has moved. Both the buy plan and allocation needed current signals, not historical ones.
What I built A buy plan tool for 1,200+ wholesale stores and an allocation tool for 400+ chain stores. Both replace fragmented, manual processes with current-data decision support.
+15%
Sales rate improvement
1,200+
Wholesale stores
400+
Chain stores
Read full case →
Product Teardown · Fintech
Razorpay
The gap between merchant sign-up and first transaction
· 12 min read  ·  Analytical proposal
The obstacle: B2B fintech has a structural activation lag. Merchants complete onboarding and go quiet before transacting. The metric that reveals this gap is also the hardest to instrument.
Approach Mapped the activation gap, proposed a 7-day north star metric, and designed a WhatsApp-first onboarding flow for Tier-2 merchants.
+34%
Activation (modelled)
20→6d
TAT (modelled)
₹4.2Cr
GMV modelled
Read full case →
Product Case Study · Consumer Fintech
PhonePe
Why 65 Cr users who trust you with money won’t buy insurance from you
· 15 min read  ·  Consumer psychology
The obstacle: Trust in payments and trust in insurance are psychologically different. PhonePe earned one kind. The product work is in understanding why the other cannot simply be inherited.
Approach Applied behavioural economics to identify three design-addressable barriers, then built a moment-based trigger framework and progressive commitment model.
65 Cr
Registered users
2.8%
India life insurance penetration
85%
Revenue from payments
Read full case →
Capabilities

What I bring.
What I am building.

What I bring on day one
Problem structuring & framingSCQA · Pyramid
Demand analytics & forecastingPower BI · SQL
Stakeholder alignmentCross-functional
Trade-off thinking & prioritisationRICE · ICE
Data storytellingExec-level comms
Financial modellingUnit economics
What I am actively working on
User research & interview designIn progress
Product spec writing (PRDs)Practising daily
Python for product analyticsIntermediate
A/B testing & experimentationFoundational
Go-to-market strategyCase-study level
How I work

Principles I
operate by.

01
Most operational problems are visibility problems first. Before you fix the process, make sure you can see the process clearly. If the data is lagging, the diagnosis is wrong.
02
Faster decisions matter more than perfect dashboards. A 7-day data cycle that people actually use beats a 1-day cycle that took six months to build and sits in a tab nobody opens.
03
Teams do not adopt systems they were not part of building. The stakeholder conversation that surfaces resistance early is not a delay. It is the work.
04
Good planning systems reduce reaction time, not just reporting effort. The goal is not a prettier spreadsheet. It is a team that makes the right call two weeks earlier.
05
Choose the metric before you build the solution. Most product failures are measurement failures. Someone tracked activity when they should have tracked outcome.

Looking for
the right role.

I reply within 24 hours.

✉ anudev374@gmail.com ↗ LinkedIn 📞 +91 62631 25926 ↓ Resume PDF