P&L and analytics systems
Full-stack profitability reporting from invoice-level data. MRP to net income, drilled by brand, SKU, channel, geography and forecast version. Built to survive a month-end close, not just a demo.
Mehul Thakkar
A Chartered Accountant who writes the code.
Thirteen years turning finance and operations problems into systems that run themselves. P&L platforms for Hindustan Unilever and Paper Boat. Workflow applications for Bajel and Swagat Housing Finance. RPA for Big-4 consulting firms.
- Chartered Accountant
- CISA
- CEH
- MOS Excel Expert
- 13+
- years building
- 1,300+
- projects delivered
- 99,000+
- hours saved for clients
- 13,000+
- people trained
- 100+
- clients served
- 20+
- cities across 4 countries
Most automation projects fail in the gap between finance and engineering.
-
01
I read the P&L
Thirteen years as a Chartered Accountant means I understand the number before I automate it. No discovery phase spent explaining what contribution margin is.
-
02
I design the system
Scoping that accounts for how finance teams actually close a month, not how a process diagram says they do.
-
03
I write the code
Python, Laravel, SQL, DAX. I build it, so the estimate is honest and the handover is real.
What I do
Custom web applications
Workflow and approval systems where a spreadsheet has stopped coping. Tender pipelines, loan origination, provisioning requests, physical stock checks. Real applications with real users and real audit trails.
Process automation and RPA
The recurring work that quietly consumes a finance team. Reconciliations, variance chasing, report distribution, document verification. Automated end to end, including the awkward parts nobody documents.
AI and LLM solutions
Language models applied where they genuinely beat a rule, and rules kept where they beat a model. Document extraction, classification, retrieval over internal knowledge. Deployed against measurable outcomes.
Data engineering at scale
When the reporting problem is really a data problem. Pipelines that move and reconcile tens of millions of rows from ERP, portals and bank statements into something a business can actually query.
Corporate training
Skillset and mindset together. Sessions built around real industry projects, so participants finish a journey rather than a tool tutorial. Online and in person.
Selected work
Hindustan Unilever
Aadhaar masking at 1 million records
- Problem
- Aadhaar cards were being submitted as images carrying personally identifiable information that had to be redacted before storage, at a volume no manual process could absorb.
- Built
- An automated pipeline that reads each card image, locates the identifying digits and masks them, processing a million records without human handling of the raw documents.
Lakme Lever
Revenue forensics across 400 salons
- Problem
- Discount and scheme coupons were being exploited at salon level and head office had no visibility to catch it while it was still happening.
- Built
- A daily-refreshing dashboard tracking scheme and coupon usage across 400 company-owned salons, flagging suspicious patterns so head office could curb exploitation rather than discover it at year end.
Bajel Projects
Tender, project and hedging workflow
- Problem
- Tenders, project schedules and commodity exposure lived in disconnected spreadsheets, leaving metal price and forex risk on approved projects effectively unmanaged.
- Built
- A web application covering the entire workflow: tenders, projects, metal prices, forex exposure and project schedules, with hedging contracts and physical inventory purchase modelled as risk mitigation.
Abbott India
Section 194R withholding tax automation
- Problem
- Withholding tax on payments to doctors only applies once each individual crosses a threshold, which becomes unmanageable by hand at PAN-India volume.
- Built
- An automation that computes the liability per doctor and carries month-on-month balances, triggering deduction as each crosses the threshold, so compliance scales without a larger team.
Kansai Nerolac
Net contribution P&L from 2 crore invoice rows
- Problem
- Product and business-level profitability could not be seen from invoice-level records at the volume the business actually generated.
- Built
- A product and business P&L built directly from invoice-level data, processing 2 crore rows through a SQL database and integrated with SAP.
Hindustan Unilever
Factory health across 28 plants
- Problem
- Machinery condition across 28 Indian factories was visible locally but invisible centrally, so problems surfaced as failures rather than trends.
- Built
- A dashboard reflecting the health of machinery across all 28 factories, pulling telemetry through Azure Data Explorer and surfacing it alongside operational data.
Swagat Housing Finance
Loan origination workflow for an NBFC
- Problem
- Loan files moved from creation to sanction through manual handoffs, with no single view of where any file stood.
- Built
- A web application managing the workflow end to end, from loan file creation through to the sanction letter, with each stage owned and visible.
Purplle
Pincode-level P&L for every SKU
- Problem
- Profitability varied sharply by delivery geography and SKU in ways aggregate reporting concealed.
- Built
- A pincode-level P&L stack covering all SKUs, alongside brand, own-brand and third-party P&L stacks and a growth analytics view identifying which brands were actually driving growth.
Bajaj Allianz
Business risk scoring across 200+ risk items
- Problem
- Risk exposure was spread across more than 200 identified risk items with no consolidated way to see concentration or trend.
- Built
- A dashboard analysing exposure across the full risk universe, with five-level drill down and trend analysis by risk item and category.
Also delivered
- Hindustan Unilever Full-stack P&L across five countries Excel / Power BI
- Hindustan Unilever GST reconciliation and HSN rate simulation Power BI / Power Query / Power Pivot / DAX / SQL
- Hindustan Unilever Payroll cost analysis Python / Power Query / Excel Macros
- Hindustan Unilever Cost centre spend analytics Python / Power Query / Excel Macros
- Hindustan Unilever Variance report automailer Excel Macros
- Hindustan Unilever Influencer campaign dashboard Power BI
- Lakme Lever 400 salon dashboards from one dataset Excel Macros
- Lakme Lever All-salon product and services P&L Power BI
- Paper Boat Three-level full-stack P&L Power BI
- Abbott India Division to SKU variance analysis Power BI / Excel / Python
- Bajel Projects Finance provisioning web application Laravel / AWS
- Bajel Projects Inventory and insurance web applications Laravel / AWS
- Lakeshore India Management Mall analytics suite Excel / Power BI
- Hindustan Petroleum Retail outlet mapping on national highways Power BI
- Bajaj Electricals SAP migration data sanity check Excel / Python
- Vedanta Cashflow reporting from ERP and Hyperion Excel Power Query
- Viacom18 Cashflow automation from bank statements Excel Power Query / Python
- Vertiv GST versus purchase register reconciliation Excel Power Query / Python
- Bridgestone OEM tyre formulation costing Excel Power Query
- Parason Machinery Raw material forecast simulation Excel / Python
- Top-10 consulting firms RPA bots and document verification Power Automate / Python
Who I have built for
Stack
Languages
Web
Cloud
BI
Spreadsheets
Data and ETL
Automation
Databases
Models
Training
Training built on real industry projects rather than tool how-tos, so participants work through an entire problem instead of memorising menus. The outcomes clients report are shorter reporting turnarounds, better reporting accuracy, more citizen-development inside their teams, and higher technology adoption generally.
- 13,000+
- students
- 2,500+
- Chartered Accountants
- 600+
- batches
- 1,500+
- lectures
Credentials
- Chartered Accountant, ICAI
- Certified Information Systems Auditor, ISACA
- Diploma in Information Systems Audit, ICAI
- Certified Ethical Hacker, EC-Council
- Microsoft Office Specialist: Excel Expert
- Certified OKR Practitioner, Levels 1 and 2
- Certified Business Coach, Evercoach
- Certified Life Coach, ICF approved
- Certified Leadership and Executive Coach, ICF approved
- Certified OD Coach, ICF approved
- NLP Practitioner, ABNLP approved
- Diploma in NLP, ABNLP approved
What clients say
-
They helped us automate our LMS reporting through an Excel-based solution that resolved critical data points, all without changing the core system.
Siddharth R Prabhu, Executive Director, Swagat Housing Finance
-
Very innovative, extremely knowledgeable, with winnable solutions.
Rajan Losalka, Head Commercial, Hindustan Petroleum
-
Great customer service. Understands your requirements first and delivers more than expected.
Nilam Solanki
-
Really innovative stuff with great execution.
Vijit Anand
-
Commendable clarity of concepts, and the ability to share them in the easiest manner possible.
Rainbow, Training participant
Let us talk about the problem
Bring the process that is costing you the most hours. A first call is thirty minutes and costs nothing.
For work that needs a team, I run Awesome Analytics.