Key Takeaways
- Legacy ETL tools create technical debt, tribal knowledge gaps, and scalability bottlenecks that stall AI and analytics initiatives.
- Cloud-native ETL with embedded data integrity closes the gap between where enterprise data lives today and where analytics needs to go.
- Democratizing data pipelines so business users (not just engineers) can act on data, is the real competitive advantage of modern integration platforms.
Most conversations about AI readiness focus on models, compute, and strategy. But in my experience working with data teams across industries, the real bottleneck is almost always the same thing: the pipeline.
Before your data can power a recommendation engine, a risk model, or an autonomous workflow, it has to move. It has to be transformed. And it has to be trusted. That last part is where most organizations are still struggling, and it’s why I was happy to join Arawan Gajajiva, Principal Architect at Matillion, for a conversation on cloud-native ETL, AI, data modernization, and what our partnership means for data teams navigating this shift today.
I’ll cover some of the biggest takeaways here, and encourage you to watch the full discussion for more insights and a demo of ETL powered by Matillion.
What’s Blocking Organizations from Delivering Data for AI and Analytics?
The honest answer is that the biggest barrier isn’t a lack of data, but rather the weight of legacy investments.
Organizations have spent years, sometimes decades, building ETL pipelines on tools designed for a different era.
Moving away from those tools is both a technical and organizational challenge. Business processes depend on those pipelines. People have learned to work around their limitations. And the perceived migration risk is enough to keep teams stuck.
Beyond that, two other challenges come up constantly:
- Data silos. Despite years of effort toward a “single version of truth,” siloed data remains stubbornly persistent, and getting those systems to speak the same language is hard.
- Tribal knowledge. This one tends to catch teams off guard. Legacy pipelines were often built by people who are no longer at the company or are nearing retirement. Nobody fully understands what each pipeline does, why transformations work the way they do, or what might break if something changes. That ambiguity makes modernization feel impossibly risky.
Arawan framed it well from the Matillion side: in a traditional on-premises architecture, scalability was always the bottleneck. Teams would provision hardware years in advance just to plan for anticipated future load. The cloud changes that equation entirely, but only if your integration layer is built to take advantage of it.
Why is AI Making Data Modernization More Urgent Than Ever?
AI has raised the stakes for data quality in a way that nothing else has. It’s not enough to have data anymore. You need data that is accurate and fit for purpose at the moment it’s needed.
The “garbage in, garbage out” principle isn’t a new concept, but AI amplifies it. A flawed dataset that produces a slightly misleading report is one kind of problem. A flawed dataset feeding an AI model that’s making operational decisions at scale is something else entirely.
The good news is that the same cloud platforms driving AI adoption also make it far more feasible to unlock and prepare data that previously sat inaccessible in legacy systems. Unstructured text, PDFs, meeting transcripts – sources that organizations knew had value but couldn’t act on – are now processable at scale.
The challenge is getting that data ready in a governed, trusted way. That’s exactly the problem a modern integration platform, paired with data integrity capabilities, is built to solve.
Real-World Use Cases for ETL Powered by Matillion
It’s one thing to talk about modernization in general terms, but another to see where it creates tangible, near-term impact. A few scenarios that warrant the need for a solution like ETL powered by Matillion come up consistently when we talk to customers.
Here are a few that Arawan and I explored during our conversation:
Ingesting data across complex, layered architectures. Many teams are building toward a medallion architecture — raw data into a bronze layer, refined into silver, curated into gold.
Matillion’s connectors library handles the full breadth of what that requires: on-premises databases, SaaS platforms like Salesforce and Workday, and even custom connectors via REST API for sources that aren’t covered out of the box. If a system has an API, you can build a pipeline to it, without writing it from scratch.
Making pipelines accessible to the people who understand the data. One of the most underappreciated advantages of a low-code environment is who it empowers. As Arawan put it:
“It’s really just that idea of democratizing data, putting the data in the hands of the people that need to use it or get value out of it. Removing that bottleneck from the data engineering team so that data engineers can work on the more interesting, more technical problems.”
A less technical analyst who knows the data and knows what they need shouldn’t have to wait in a queue. At the same time, a senior engineer who wants to write SQL or Python can do exactly that. The platform supports both, which means teams stop being the bottleneck for each other.
Cutting costs tied to legacy infrastructure sizing. Legacy data infrastructure is typically sized for peak load, meaning organizations are paying for the biggest day of the year, every day of the year.
Cloud-native ETL changes that model. With ETL powered by Matillion and the Data Integration service of the Precisely Data Integrity Suite, you pay for what you actually use. For teams with seasonal data volumes (Q4 spikes, campaign surges, fiscal year-end reporting), that shift alone can represent meaningful savings.
Freeing up engineering capacity for higher-value work. When a business analyst can build and iterate on a pipeline without filing a ticket, engineers get their time back for work that truly requires their expertise. That’s both an efficiency gain and a cultural shift in how data teams operate and where they create value.
Power Data Modernization with Precisely and Matillion
The Precisely and Matillion partnership is built to reduce the tools you need in your stack.
Matillion is purpose-built for the cloud data warehouse world, best-in-class at transformations, scalability, and modern ETL/ELT patterns. Precisely brings what cloud-native tools typically can’t reach: enterprise-grade data quality, governance, and enrichment, embedded from the start of the pipeline rather than bolted on at the end.
Together, the result is a single, integrated platform where data moves, transforms, and arrives trusted, without the overhead of managing multiple vendors, contracts, and context switches.
As Arawan put it:
“The power is in your hands, with Precisely and Matillion. You’ve got one suite of products that addresses every layer of your AI journey — keeping it very simple, democratizing the data — and it’s going to scale with you as your needs scale.”
That’s the goal. Not just faster pipelines, but data that’s accessible to the people who need it most.
To see a live demo of ETL powered by Matillion in action, including how pipelines are built visually within the Precisely Data Integrity Suite, watch the full webinar: Cloud-Native ETL with Built-in Data Integrity: Introducing ETL Powered by Matillion.
