Survey reveals “massive productivity drain” in data engineering

Complexity, scalability and compatibility remain challenging - 70% data workers struggle with pipeline management.

  • Tuesday, 22nd April 2025 Posted 1 hour ago in by Phil Alsop

Managing data pipelines continues to present significant challenges for organizations, with an overwhelming 70% of respondents rating pipeline management as 'somewhat' or 'extremely' complex according to a recent data integration and AI report.

According to the Data Integration and AI-Readiness Survey commissioned by intelligent data integration platform Matillion, more than two thirds of respondents are struggling with pipeline management.

The survey revealed an alarming picture of data integration, an industry facing ever-increasing demand thanks to the growing business need for artificial intelligence (AI) and generative intelligence (GenAI).

Matillion CEO Matthew Scullion said: “Data engineering is boring, gritty and repetitive. Data teams are wasting valuable hours on low-value build, maintenance and management. Instead this time could and should be spent building valuable data products that can be used for new business impacting initiatives.“

The survey highlighted a substantial productivity drain with 64% of organizations reporting that their data teams spent more than 50% of their time working on repetitive or manual tasks.

Rather than increasing staff (and increasing the overheads associated with that), organizations need to identify data integration solutions that empower their current teams to work more efficiently.

Scalability emerges as another critical concern, with 89% of organizations noting issues with their current data engineering platform’s ability to scale pipelines to meet data processing needs.

Scullion added: “This survey highlights the need for a unified solution that is accessible to a wider scope of the organization - easing the load on data engineers and data teams, while empowering business leaders to make data-driven decisions. Bringing AI into this mix creates an incredibly compelling vision of data engineering of the future, where data engineers are able to focus their time on innovation and doing what they do best - solving business problems, rather than manual, laborious pipeline management.”

This Data Integration and AI-Readiness Survey was conducted in partnership with Perspectus Global in January 2025. Respondents included 307 data decision-makers and data user titles, based across the UK and the US.

Cybersecurity strategies are failing

Posted 1 hour ago by Phil Alsop
Cyber firm pleads with enterprises to wake up to the data security crisis before financial and legal fallout becomes catastrophic.
Avanade is unveiling the Avanade Intelligent Garden at this year's RHS Chelsea Flower Show in celebration of its 25th anniversary.

AI agents break cover

Posted 4 days ago by Phil Alsop
In a global survey of IT leaders, Cloudera found that enterprises are keen on AI agents, but fears around data privacy, integration, and data quality...
Economist Impact is pleased to announce the inaugural AI Compute summit, scheduled for May 22nd 2025, at the Scandic Copenhagen in Copenhagen. This...

Majority of AI projects don't make it to market

Posted 5 days ago by Phil Alsop
SS&C Technologies Holdings has published findings from a new survey: governance, process orchestration and strategic planning are critical to...

Security and compliance risks make VPNs obsolete

Posted 5 days ago by Phil Alsop
Zscaler has published the Zscaler ThreatLabz 2025 VPN Risk Report, commissioned by Cybersecurity Insiders, which highlights the widespread security,...

AI tops tech growth charts

Posted 1 week ago by Phil Alsop
Despite high interest rates, economic slowdown, stricter regulations on big tech and AI, Trump's tariff policies, and global trade wars, which hit...

94% increase in network malware

Posted 1 week ago by Phil Alsop
Other key findings show an increase in crypto miner detections, a spike in zero-day malware, a drop in endpoint malware, a rise in Linux-based...