Customers

ROQT

One workspace, 20+ clients: a consultancy running on Nekt

ROQT customer story on Nekt.

About

ROQT | Data & AI is a data and AI consultancy from Maringá, Brazil, serving more than 350 companies across the country.

Company size

Industry

Data & AI consultancy

Results

79
data sources across 23 connectors
334
data transformations
20+
end clients in one workspace

The challenge

ROQT came to Nekt with a bold idea: use the platform as the backbone of the entire consulting operation. Not for one client, for all of them. E-commerce brands, construction firms, logistics companies, football clubs, publishers and agricultural cooperatives, each with its own systems, all running on one platform without losing the essence of each business.

What they built

A consultancy, running on one platform

ROQT | Data & AI is a data and AI consultancy from Maringá, Brazil, serving more than 350 companies nationwide. It came to Nekt with a bold idea: make the platform the backbone of the whole consulting operation, not for a single client, but for all of them. In five months, ROQT built something few expected: 79 data sources across 23 connector types, 334 transformations, and more than 20 end clients running in a single Nekt workspace, from e-commerce brands and construction firms to logistics companies, football clubs, publishers and agricultural cooperatives.

What ROQT built on Nekt

  • A marketplace unification layer that consolidates orders from seven Brazilian e-commerce channels into a single schema, with SKU alias resolution, cost lookup by date and proportional commission allocation.
  • Eight separate ERP instances consolidated into unified financial models, so finance reads one number across very different businesses.
  • More than 20 clients isolated in one workspace, each with its own sources, rules and governed space.
  • Agents that query with governance. Through Nekt's MCP, all of it can be queried by AI agents, with governance and at scale, instead of client-by-client spreadsheets.

The systems behind it

Seventy-nine data sources across 23 connector types feed the workspace. A representative slice:

SystemsWhat they power
Mercado Livre, Shopee, Magalu, Amazon, Olist, Shopify, TikTok ShopSeven e-commerce channels unified into one order schema
Sienge, Omie, Oracle, F360, PrevisionERPs consolidated into unified financial models
Google, Meta, TikTok, Amazon and Shopee AdsPaid media across client accounts
RD Station (Marketing, CRM), MeetimeLeads, CRM and sales engagement
Microsoft SQL Server, Postgres, Oracle, DatabricksClient databases and warehouses
SharePoint, SFTP, REST APIFiles and custom integrations
GA4, LinkedIn, PinterestWeb and social performance

Speed is the surprise

Five months from the first connection to 380 active pipelines. That only happens when a team understands data architecture deeply and the platform gets out of the way. For a consultancy, this changes the model: what you build for one client becomes infrastructure to serve the next, faster and more consistently, without losing the essence of each business. It is the same principle behind context engineering for AI agents and a governed MCP server for data. Learn more about ROQT.