● Analytics · Excel / VBA · 15 Years

Pricing &
Promotions
Engine

A proprietary financial and analytical system built from scratch inside a major wine import company. Replaced manual, disconnected processes with a unified, formula-driven toolkit, bringing pricing discipline, promotion analysis, and supplier reporting under one roof for the first time.

MB Group · Moscow, Russia · 2009–2024 · 15,000+ SKUs · 40–70 supplier reports / month

15,000+
SKUs managed
70/mo
supplier reports
15
years in production
0→1
built from scratch
the problem

Before this system existed, the company had no unified tool for pricing or promotion analysis. Every decision required manual lookups across disconnected files. Monthly supplier reports, up to 70 per cycle, were each built from scratch by pulling data from 1C (ERP), reformatting it, and populating individual Excel files by hand. The process took 2 full weeks every month and left no time for actual analysis.

There was no way to quickly answer: Is this promotion profitable? What's our margin after the retailer retro-bonus? How are depletions trending across channels?

the solution: two modules
Module 01

Pricing & Promotion
Financial Model

A multi-layered Excel financial model covering every cost variable in the wine import P&L. For each SKU and each promotion scenario, the model computed the full margin stack in real time.

PRICE (base price list, RUB) 100%
− Cost of goods EUR × rate
− Retailer retro-bonus 3% – 12%
− Cash bonus / KU varies
− Manager bonus fixed %
− General overhead (OHZ) % of price
− Logistics per bottle
− Promo mechanics 1+1, 1+2…
= NET MARGIN % real-time
●Auto-pulled base prices and EUR costs from 4 linked reference files
●Separate promo block for free-bottle mechanics (1+1, 1+2, etc.)
●Built-in minimum price calculator for 5% and 7% target margins
●Real-time validation: flagged "invalid price" or "increase price" when margin went negative
●"Send to client" export sheet: auto-generated promo summary for retail partners
Connected to: [1] Price list [2] Nomenclature DB [3] SKU cost reference [4] Sales by SKU
Module 02

Unified Supplier
Reporting System

A centralized Excel database that aggregated all data exports from 1C (the company's ERP) and fed 40–70 individual supplier reports semi-automatically each month.

Before
● Raw 1C exports, reformatted manually per report
● Each report built from scratch every month
● ~2 weeks per reporting cycle
● Consistency varies, human error risk
● No time left for analysis
→
2 weeks → 3 days
After
● Single unified base, updated once from 1C
● Formulas pre-linked: refresh = update
● ~3 days per reporting cycle
● Standardized across all suppliers
● Channel trends, YoY, import vs depletion
Data architecture
Sales by channel Stock levels Imports (btl + EUR) Goods in transit Invoices
1C ERP
exports
→
Unified
Excel Base
→
40–70
Supplier Reports
●SKU-level price history (EXW by year, with and without budget)
●Monthly depletions Jan–Dec (bottles + 9L cases)
●YoY growth % per SKU group
●Channel breakdown: ON/OFF trade, St. Petersburg, Regions, Private
●France, Italy, Spain, Chile, Argentina, and other origins
impact
15,000+
SKUs managed
Across all pricing, promo, and reporting tools simultaneously
70/mo
Supplier reports / month
Generated semi-automatically. France, Italy, Spain, Chile, Argentina and more
2 wks
→ 3 days
Reporting cycle
Monthly reporting time reduced. Time freed for actual business analysis
15 yrs
Years in production
Actively used 2009–2024. No external vendor, no license fee, no replacement needed
why this was hard to build
Scale
15,000+ SKUs, no room for error

The price list alone covered thousands of SKUs across dozens of suppliers and countries. Every formula had to work correctly at full scale: a single broken reference cascaded across hundreds of rows.

Interconnectedness
4 live external data sources

The financial model pulled live data from 4 separate reference files simultaneously. Any structural change in one file could break downstream calculations. The architecture had to be resilient to routine updates.

Business logic depth
8+ cost layers, real-time validation

Promotion profitability calculations accounted for 8+ simultaneous cost layers, including VAT-adjusted KU, multi-type retro bonuses, and mechanic-specific free goods costs, while enforcing margin floors and flagging invalid pricing in real time.

Reporting diversity
Every supplier is different

Each supplier had unique SKU groupings, price tiers, and channel structures. The unified base had to normalize all of this into consistent output formats, without hardcoding supplier-specific logic that would break at scale.

built with
Core Tools
Excel / VBA Financial Modeling Data Normalization
Architecture
Multi-file VLOOKUP Cross-workbook Linking 1C Integration

need analytics
or a reporting system
for your business?

get in touch ↗ back to portfolio