Fivetran, a worldwide pioneer in the development of data infrastructure for the utilization of artificial intelligence, has rolled out the Open Data Infrastructure (ODI) Data Access Benchmark. The benchmark is the starting point of a series of benchmarks to expose the operations of large software vendors with regards to data access for their customers. The transition from experimental to full-scale production of AI by the majority of companies have in fact resulted in a new bottleneck.
This bottleneck is the very type of software vendors who have set the data of the customers are actually the ones who impose the barriers both technical and commercial. Some of these barriers are incomplete APIs that hide key data objects, throughput throttling that limits the speed of the real-time replication, and pricing models that are very aggressive such as egress fees that are basically penalizing data moving.
Also Read: InfluxData and Litmus Forge Strategic Partnership to Accelerate Scalable Industrial AI at the Edge
The ODI Data Access Benchmark provides a transparent, standardized framework to measure these constraints. By evaluating vendors on coverage, performance, and cost, the benchmark helps organizations determine if their current technology stack can actually support the high-velocity demands of modern AI.
“What we’re seeing is a shift where access to data is becoming more restricted, both technically and commercially,” said George Fraser, CEO of Fivetran. “Oftentimes, those constraints show up as hidden costs or dependencies that push companies toward specific walled-garden ecosystems. As AI becomes more central to how companies operate, organizations need an open data infrastructure that gives them the flexibility to access, move, and use their data without lock-in.”
Breaking Down the Barrier to AI Scalability
Modern AI workloads require unfettered access to high-quality, large-scale datasets. However, many enterprise platforms utilize “walled garden” strategies that keep data siloed. Fivetran’s new benchmark evaluates vendors across three critical pillars:
- Data Coverage: Assessing whether core data objects are programmatically accessible or hidden behind proprietary walls.
- Performance & Speed: Measuring support for high-throughput replication and advanced methods like Change Data Capture (CDC).
- Economic Transparency: Identifying whether data can be moved without punitive egress charges or restrictive licensing terms.
The benchmark arrives at a critical juncture. Fivetran’s recent research indicates that nearly 97% of tech leaders have seen their AI or analytics programs hampered by infrastructure issues. By making this data publicly available at OpenDataInfrastructure.com, Fivetran aims to empower businesses to make informed procurement decisions that prioritize data sovereignty.
A Call for Open Data Infrastructure
The message here from Fivetran is that there should be a move toward “Open Data Infrastructure,” which would include an ecosystem based on interoperability and data portability as a matter of course. The intention is to grow the benchmarking program, including new vendors and data sources, to keep the industry accountable.
As a platform that manages data movement for thousands of global organizations including LVMH, Pfizer, and OpenAI-Fivetran possesses unique visibility into how these vendor models impact operational efficiency. The ratings provided in the benchmark are derived from publicly available documentation and reflect generally available capabilities, ensuring a fair and consistent evaluation for all parties involved.
