Katana Turbocharges AI and Analytics Capabilities with a Smart Data Platform

Katana’s success is rooted in its ability to give growing businesses what they have long lacked: a real-time, unified source of truth for inventory management. By consolidating fragmented data from e-commerce, accounting and production systems into a single, coherent hub, Katana enables companies to track stock levels accurately, respond faster to demand and make confident, data-driven decisions. This capability is powered by Aiven for Apache Kafka and Aiven for ClickHouse, which together transformed Katana’s data infrastructure to deliver instant analytics and seamless performance at scale. The impact is measurable. Customers see stronger inventory turnover and significant year-over-year sales growth. With Aiven’s recent ClickHouse revamp introducing a more powerful architecture, object storage integration and flexible scaling, Katana now benefits from even faster dashboards, streamlined DevOps and AI-assisted development. The company is well positioned to lead the next generation of high-performance inventory management.

From start-up to inventory management leader with plans for more growth

In just seven years since its initial investment, Katana has become an award-winning company with more than 1,500 customers in 70 countries. The company provides an end-to-end inventory management platform that connects purchasing, production, warehousing and sales in a single, real-time platform. From its initial focus on small and mid-size manufacturers, Katana has now expanded its reach, adding fast-growing e-commerce and retail SMBs.

These modern merchants are seeking smarter, more agile inventory control. They sell physical products across multiple channels and storefronts, yet often find themselves caught between the limitations of spreadsheets and the complexity of larger-scale ERP systems when managing fast-moving inventory. Many rely on disconnected tools such as Shopify and QuickBooks, and have multiple channel specific shopfronts, leaving data fragmented and visibility limited. By filling that gap, Katana synchronizes data across all sales and operational systems in real time, giving merchants a unified source of truth.

Katana now supports customers managing more than USD 3 billion in gross merchandise value annually, with plans to support even more businesses with the tools and resources they need to flourish.

Expanding customer base changes approach to data infrastructure

Katana’s real-time master-planning engine optimizes production and prevents stockouts by reacting instantly to incoming multi-channel sales data. To achieve this, Katana manages and stores high volumes of data related to inventory, sales and transactions, and its value proposition is highly dependent on real-time data synchronization, system reliability and scalability.

As it expands beyond manufacturing into high-volume, multi-channel commerce, its data infrastructure needs to evolve significantly. Whereas traditional manufacturers typically handle smaller numbers of high-value sales, modern merchants are in the volume business as they handle vastly more transactions of lower average value at far greater speed.

“It takes a different technical approach,” says Valter Kungla, Senior Engineering Manager at Katana. “For our customers, it’s all about real-time inventory management. They can’t afford to sell a product through one sales channel that has already sold out on another. So we needed more focus on performance and concurrency. In short, we needed to strengthen and evolve our own data infrastructure to support modern merchants’ real-time, high-volume, multi-channel commerce.”

A distributed data environment of integrations and microservices

Customers connect the Katana platform to different third-party solutions and sales channels, notably book-keeping, accounting and e-commerce platforms, leaving data spread across multiple systems. In addition, Katana’s own architecture is built on around 150 microservices, each responsible for its own data, adding further complexity to the overall data landscape.

The company was using Amazon Aurora RDS for PostgreSQL as its main analytical and operational data store, with all queries for insights dashboards and reporting running directly against RDS. But as the company grew and expanded its services, queries were slowing down, especially for complex analytics and reporting which could take up to an hour to complete. “It was causing noticeable delays in customer dashboards,” says Kungla. “That’s not what we wanted to be known for.”

Alongside Amazon RDS, Katana was also using Aiven for Apache Kafka® as its data pipeline backbone. Aiven for Kafka enabled a Change Data Capture (CDC) stream that reflected database updates and acted as the source of truth for real-time data movement between services. All data events, such as inventory updates or sales transactions, originated or flowed through Kafka.

Aiven for ClickHouse aligns perfectly with requirements

Katana began its search for a more robust, more scalable system that could handle fast transactions in real time and run seamlessly on AWS. It commissioned independent research that identified ClickHouse as the best database for its analytics and data aggregation needs.

A fast, open-source, column-oriented database management system, ClickHouse enables users to generate analytical data reports in real time with SQL queries. Katana wanted it to provide a unified analytics layer that consolidates data from all its internal microservices and external integrations. This would give customers a single, consistent overview of their operations within Katana, even though the underlying inventory, sales, or production data was coming from many different sources.

Aiven for ClickHouse aligned perfectly with Katana’s choice and AWS setup. “Aiven for Kafka had proved to be exceptionally reliable, with no disruptions or downtime. So Aiven for Clickhouse running on AWS was a natural fit, giving us a scalable, high-performance analytics layer with simplified deployment and management,” says Kungla

Aiven for Startups delivers additional benefits

In addition, Katana could take advantage of the Aiven startup program, through which it benefited from USD 100,000 in Aiven credit and high-touch technical guidance.

The credits allowed Katana to use and evaluate Aiven for ClickHouse and access Aiven engineers for almost a year without making any upfront financial commitments. “As the ClickHouse architecture is different from RDS, the transition was a learning curve for our team and required careful setup,” says Kungla. “Aiven engineers were very helpful during the setup of Aiven for Clickhouse. And the results are absolutely worth it.”

To move to ClickHouse, the project team had to completely rewrite roughly 30 complex SQL queries, many of which involved multiple joins and aggregations. As an AI-first engineering organization, Katana used AI coding tools like Cloud Code and large language models to automate 85% of the query conversion work from PostgreSQL to ClickHouse syntax. With only minor manual corrections required, the AI-driven conversion dramatically accelerated the migration process.

Katana also leveraged internal AI agents to automate the generation and validation of complex ClickHouse queries. This AI-assisted approach accelerated integration work dramatically — reducing development effort by roughly 80% compared to previous manual processes.

Performance improvements enhance customer and engineering experience

Today ClickHouse enables Katana to aggregate, query and present a coherent, unified view of data, both for internal use and for customers, through the Katana platform’s insights dashboards. Queries that once took up to an hour can now run in seconds, delivering insights in real time or near-real time. This architecture provides customers with faster, more intuitive dashboards and clearer visibility into inventory performance, significantly improving their day-to-day experience and supporting Katana’s reputation for reliability as the company scales.

These performance gains also translate into measurable business impact: on average, customers using Katana report a 21% increase in inventory turnover and a 60% increase in year-over-year sales, reflecting the value of real-time visibility and smarter, more responsive planning.

Recently, Aiven for ClickHouse underwent a major revamp that has amplified these gains even further. The new architecture delivers dramatically improved query performance, greater scalability, and even better price efficiency, empowering Katana to do more with less. By introducing features like Dynamic Disk Sizing and object storage integration, Aiven has made ClickHouse both more powerful and more cost-effective, enabling Katana to handle ever-growing data volumes without overprovisioning or compromising on speed.

With Aiven ensuring seamless operation across the Aiven Platform, Katana can dedicate its energy to innovation and delivering business value. Freed from infrastructure management, Katana can redirect resources toward product development, and rather than maintaining a large DevOps team, it has a smaller, more focused engineering team.

“Aiven has long been a mission-critical partner for us. Kafka is the source of truth for our data, and everything depends on it. Adding ClickHouse has taken it a step further. Everything just works. That kind of reliability is rare, and Aiven delivers it effortlessly. We’re very happy to have a partner we can trust,” says Kungla.

Ready for richer analytics and AI-powered services

Implementing Aiven for ClickHouse is a major milestone in Katana’s data evolution toward a unified, scalable data infrastructure and a customer offer based on richer analytics and more integrations. ClickHouse now serves as the central platform for aggregating and analyzing data, unlocking many new possibilities including plans to integrate more AI support into customer-facing services.

“With Aiven’s support and tooling, we know we can scale as needed while keeping our focus on business priorities. Together, we’ll continue building the best data infrastructure to support Katana as we grow and evolve,” says Kungla.