/ Case study /

Research Data Hub Platform

One Platform to Unify Them All: Building the Research Data Hub That Powers a Global Market Research Operation

$80M

In annual revenue supported by the research metadata the platform manages
$45M

In additional annual revenue supported by the downstream cloud analytics platform it feeds
1

Single source of truth for all research metadata — programs, brands, product lists, and survey waves across dozens of countries

/ 01 / Executive summary

Executive Summary

Primotly partners with a global market research leader on the ongoing development and performance optimization of their central research data hub — the internal operational platform that underpins their entire research lifecycle. The platform serves as the single source of truth for all research metadata: programs, brands, product lists, and survey waves spanning dozens of countries and languages. Research staff input and manage data through structured Excel-based workflows, which the platform ingests, standardizes, and makes available to downstream analytical teams. A major ongoing initiative — project FLOW — is systematically refactoring and re-engineering the platform's backend to address accumulated performance bottlenecks and code quality issues. The platform supports approximately $80 million in annual revenue.

/ 02 / About the Client

About the Client

The client is a global enterprise specializing in advanced market research, consumer behavior studies, and brand tracking. They operate in dozens of countries, running hundreds of simultaneous research programs for their own corporate clients across FMCG, technology, healthcare, and other sectors. Internally, their research operations teams — program managers, brand specialists, and data coordinators — rely on a centralized data hub to manage the metadata that drives all research activity. (All details have been anonymized due to NDA.)

Services

Team Extension
Web Development
Performance Engineering
Legacy Code Refactoring

Industry

Market Research

Collaboration Model

Team Extension — Long-Term Dedicated Partnership

Team Size

2 people (1 Frontend Developer, 1 Backend Developer)

/ 03 / The challenge

The Challenge: A Growing Operation Straining Against an Aging Platform

The research data hub is a mature, business-critical system. Over years of growth, it accumulated the technical debt common to large, long-running enterprise platforms: performance degradation, code quality issues, and an architecture that made new feature development slower and riskier than it should be. The consequences were concrete — slower internal workflows, higher risk of bugs during development, and a platform that required careful handling rather than confident iteration.

01/

Performance Bottlenecks

Key platform operations had become significantly slower over time. Response times for data-heavy workflows were impacting the productivity of internal research teams who used the platform daily.
02/

Legacy Code Quality

Years of accumulated development had produced sections of the codebase with poor structure, limited test coverage, and high coupling — making changes difficult and regressions more likely.
03/

Excel as the Foundation

Research data entry is structured around Excel — staff upload Excel files that the platform ingests, validates, and standardizes. This workflow is deeply embedded in the organization's operations and will remain so. The challenge is not to replace it, but to ensure the platform handles it reliably, efficiently, and at scale.
04/

Data Consistency Across Sources

Research data flows into the hub from multiple internal sources and teams, each with their own formats and conventions. Ensuring that downstream analytical teams always work with consistent, trustworthy data is a continuous challenge.

/ 04 / The Primotly solution

The Primotly Solution: Performance Engineering and Systematic Refactoring

Primotly's work on this platform falls into two complementary tracks: ongoing feature development to meet evolving business requirements, and a systematic performance and quality improvement initiative known internally as project FLOW.

Key Capabilities Maintained and Developed

Program Management

Research staff create and configure research programs, defining their geographic scope (countries and languages), ownership, and lifecycle. The platform enforces a structured hierarchy that replaces ad-hoc spreadsheet organization.
Brand and List Management (Wave List Manager)

A central registry of brands, product lists, and SKUs used across all research programs. Brand configurations — including market activation, localization settings, and classification tags — are maintained here. Changes submitted by external corporate clients via the self-service portal are reviewed and approved by staff before flowing into this module.
Wave Management

Research waves (defined time periods for data collection) are configured and tracked centrally. Teams manage wave schedules, scope, and parameters across multiple programs simultaneously.
Excel-Based Data Ingestion

Staff upload research data via structured Excel files — the established and continued standard for data input across the organization. The platform ingests, parses, and standardizes these files, making the data available in a consistent structure for downstream analytical systems.
Data Quality Harness (in development)

A major initiative to define a universal metadata schema that all data entering the platform must conform to. The goal is to ensure that every analytical team — regardless of which source their data came from — works with data in a consistent, trustworthy structure.
Downstream Integration

Standardized data is exported to the analytical infrastructure, including the cloud analytics platform that supports ~$45M in additional annual revenue.

/ 05 / Technical deep dive

Technical Deep Dive

01/
Project FLOW: Systematic Performance and Code Quality Recovery
- The Problem: The platform had accumulated significant performance debt. Backend operations that internal teams depended on daily had become unacceptably slow. Root cause analysis revealed a combination of inefficient query patterns, high-coupling code structures, and legacy sections of the codebase that had never been properly refactored.
- The Solution: Primotly launched project FLOW — a dedicated refactoring and performance engineering initiative running in parallel with feature development. The approach is systematic: identify the highest-impact bottlenecks, refactor the underlying code structures to eliminate coupling and improve testability, then optimize query and processing patterns. Each FLOW sprint produces measurable backend speed improvements.
- The Result: measurable reduction in backend response times for the platform's most critical operations, with improvements delivered incrementally across FLOW sprints.
02/
Reliable Data Standardization Across Multiple Input Sources
- The Problem: Research data enters the platform from multiple internal teams and sources, each with their own Excel formats, naming conventions, and data structures. The platform must absorb this heterogeneity and produce consistent, well-structured data for downstream analytical teams. Inconsistencies that slip through create problems at the analysis stage.
- The Solution: The platform applies standardization and validation rules at the ingestion layer. Uploaded Excel files are parsed and each data point is checked against program- and wave-specific rules before being committed. The data quality harness initiative will extend this further by defining a universal metadata contract that all inputs must satisfy, regardless of source.
- The Result: downstream analytical teams receive data from a single, consistent source — regardless of how many different internal teams contributed to it. This is the core value the platform delivers: trusted, standardized data as the foundation for all analysis.
03/
Maintaining a Large-Scale Monolithic Platform Under Active Development
- The Problem: The platform is a large monolith that has grown over many years. Without discipline, monolithic architectures of this scale degrade into tightly coupled codebases where every change carries significant risk.
- The Solution: Primotly applies consistent architectural standards across the codebase — clear module boundaries, separation of concerns, high test coverage on refactored sections, and rigorous code review. The FLOW initiative specifically targets high-coupling sections and refactors them to cleaner patterns before they become development blockers.
- The Result: a more maintainable, lower-risk codebase. Each FLOW sprint reduces both performance debt and the complexity cost of future development.

/ 06 / Business impact & results

Business Impact & Results

The platform is the data backbone of a global research operation — every analytical team depends on the consistency and reliability of the data it provides.

$80M Revenue Supported

The platform manages research metadata for programs generating approximately $80 million in annual revenue — making its performance and reliability directly tied to business outcomes.
Performance Improvements via FLOW

Project FLOW delivers measurable backend speed improvements in each sprint, directly improving the daily productivity of internal research teams who depend on the platform.
Consistent Data for Analytical Teams

By standardizing data ingested from multiple internal sources, the platform ensures that analytical teams across the organization work with consistent, trustworthy data — regardless of where it originated.
Foundation for Broader Analytics

The platform feeds data to the cloud analytics platform that independently supports ~$45M in annual revenue and is growing rapidly.
Long-Term Strategic Partnership

The client has committed to a multi-year development roadmap with Primotly. The data quality harness alone is scoped as a several-year initiative — a signal of the platform's ongoing strategic importance.