Matthew Scullion, is the CEO and founder of Matillion – the leading provider of data transformation software for cloud data warehouses.
The company has reinvented data transformation in the cloud era – as the only provider enabling true data transformation in the cloud. Its unique ELT (vs ETL) approach leverages the power of an enterprise’s chosen cloud data warehouse for data transformation, maximizing scalability, speed and savings.
Matillion’s 800+ customers include Accenture, Amazon, Bose, Docusign, Fox, GE, Merck, Siemens, Sony, Splunk, VIstaprint, and Zapier.
Matthew has over 20 years of experience working at the intersection of business and technology, with the majority of his career as a software and IT professional.
You’re a lifelong entrepreneur having founded your first company when you were 18. What was this first company? Looking back was there something that could have been done differently to enhance the odds of success?
It was a little company called Real Information Systems – we did groupware development and implementation. Whilst I was there as an evenings-and-weekend project I built an early web content management system which we managed to sell to some decent customers and which won a few awards. At one of the award ceremonies we met a large American firm who shortly thereafter bought the company.
And what would I have done differently? Pretty much everything. I was actually 17 when I got involved in that business and my partner was a great guy (and we’re still friends) but not much of a businessman. If I had the skills set I have now, back then, I think we’d have had a much more successful business! Mostly I learned about people skills – everything in business (or knowledge business like software) is down to the quality of the people – getting them in the right seats at the right time – and getting them aligned, and in great culture.
After this first venture you then worked in commercial IT and software development for 15 years at a number of British and European systems integrators. What were some of these roles?
I worked up through the management of developers, some pre-sales through to P&L responsibility for the software division of a large European SI… this gave me a lot of customer experience, particularly in larger customers. And I ran BI, Software, SDLC, and Integration practices – which gave me all the tools I needed for Matillion.
Was there a gap in the marketplace that you noticed in these different types of roles that inspired you to launch Matillion?
I’d seen at this previous firm the power of well-implemented BI. And what a heavy lift it was to get it right. I’d also seen a German company that had been really successful with a more prepacked, industry vertical BI solution for mid-sized businesses. Finally, we’d been working with AWS and could see the writing on the wall for on-prem solutions. So cloud-based, turn-key, fully managed BI seemed like a gap that we could fill. That’s not what we do now, but it’s what we did at the beginning and what led us to build Matillion ETL years later, which is now what the business is all about.
After launching Matillion in 2011, you convinced three of your former bosses to invest. How did this discussion go, and what was it about the idea that convinced them to get on-board?
I decided to quit my job and set up the company whilst on holiday in Spain with my wife. I’d been having a rough time and the previous place and she encouraged me to do it. I went home and put a business plan together which I then pitched to these 3 people.
I think (in fact know) that they thought I didn’t know too much more than them about the cloud and BI. But they’ve subsequently told me that they weren’t backing the business plan – rather they were backing me, as they thought I had a chance of making a success of something.
I guess that was off the back of the 10+ years I’d worked with and for them before, and whilst it’s always a good idea to be always thinking about your integrity.
How would you describe Matillion in as few words as possible?
Matillion is data transformation for cloud data warehouses.
One of the key differentiators between Matillion and other solutions is Matillion’s ELT (extract, load, transform) approach versus the more commonly used ETL (Extract, transform, load). Could you describe these two approaches and what makes ELT superior?
When data integration products entered the market years ago, they delivered three areas of value still required today: 1) extract, 2) load (get data from source systems into the data warehouse) and 3) data transformation (join together siloed data, denormalise, add value and embellish business logic and metrics). In other words, early products made enterprise data “analytics-ready,” in a complex IT environment, at scale and as a team.
However, these pre-cloud, legacy products were not built for the cloud. So, they were slow, difficult to manage and scale, hard to procure, and unable to take advantage of the underlying power and features of cloud platforms.
At this time – before 2015 – Matillion was focused on building and managing data warehouses for its customers in the cloud. We grew frustrated by the lack of products available to help support our clients’ businesses efficiently and at scale. We needed a product that would deliver simplicity, speed, scale and savings, but one didn’t exist. So, we built it ourselves. About five years ago, we launched Matillion ETL (extract, load, transform), a tool built specifically for cloud data warehouses (CDWs) including Amazon Redshift, Google BigQuery, and Snowflake
Data from disparate sources can be easily collected, but simply loading it into a central place inside a CDW does not yield high-quality data insights. Data must be transformed before it can be used with an analytics tool, or any downstream process such as machine learning. The inability to transform data contradicts the purpose of a data warehouse in the first place, leaving customers unable to build and manage complex data models, or get high-quality analysis.
Most solutions today still extract data, transform it into raw data that’s suitable for reporting and analytics, and load it into a target platform (eg a relational database). The ETL engine is, therefore, a compute resource, and must be powerful (read: expensive) enough to handle large amounts of data that will be transformed. Additionally, the environments running ETL software are not built to scale the same way as data warehouses. Therefore, as data volumes increase, these environments consume more IT resources, create bottlenecks in the data chain and can negatively impact an enterprise’s reporting and analytics. This can lead to bad business decisions made slowly, resulting in missed opportunities.
Our similar but different extract-load-transform (ELT) approach offers increased performance and value by handling the extract and load in one move, straight to an enterprise’s target data platform, using the power of a CDW’s processes to perform the transformations once loaded. Pushing transformations to the data warehouse itself requires only one powerful piece of infrastructure. Further advancements on relational databases make transforming data in-database easier, faster and more cost-efficient. This results in savings on infrastructure, better performing workloads, and shorter development cycles. An enterprise’s data is quickly migrated and immediately available for transformations and analysis based on current business challenges and needs.
If ETL looks to improve its biggest setback – performance – Matillion expects the result to be an ELT workflow. That’s the best way for an enterprise to leverage its data efficiently, cost-effectively, and for fruitful analysis to drive business growth.
Matillion has created the Matillion Data Loader which is a free SaaS-based data integration tool. Could you describe this tool in detail and what makes it so user friendly?
Today, businesses have new needs around data. The technology landscape is changing. IT data professionals still need enterprise-scale and powerful transformation capability, and always will – but at companies that compete using data, there is a new type of data user that doesn’t necessarily work in IT. These users may work in marketing, finance, and other disciplines across the business. They want to use data to find answers to their business questions. Matillion Data Loader addresses these users, and supports businesses at earlier stages in their cloud data journey. The product advances the speed and simplicity of data integration for those data users. It’s a lightweight SaaS solution that enables data analytics professionals and business users to easily migrate data with a powerful and scalable product. Customers can gradually upgrade to more sophistication and depth (Matillion ETL) as needed.
We offer Matillion Data Loader free of charge. This has disrupted the market, and demonstrates our commitment to meeting and serving enterprises wherever they are on their cloud data journey, helping them to effectively access data for intelligent business decisions.
Matillion enables enterprises to tap into the full power of cloud data to make data transformation easier, faster, and more cost-effective. Can you describe some of these functionalities?
Matillion ETL is a cloud tool that helps businesses of all sizes, across industries, with enterprise-grade, sophisticated data transformation to drive their analytics at scale on Amazon Redshift, Snowflake, and Google BigQuery. Our products can join, filter, rank, convert, aggregate, and transpose data from various sources to ease cloud migrations and transformations. They connect to platforms like Amazon S3 and Amazon RDS, Google Analytics, customer relationship management portals like Salesforce, social media platforms (Facebook and Twitter), and even payments processors (Stripe, Paypal), and they enable users with the right permissions to import and export projects or suss out the status of ongoing tasks from instance-level audit logs.
Moreover, Matillion enables admins to design and schedule reusable, parameter-driven data jobs and visually orchestrate workflows with transactions, decisions, and loops, in part through Bash and Python scripting and component generation tooling. Perhaps more useful still, it lets them generate a downloadable report that details layout, components, SQL, and properties for any given job with a single click.
Is there anything else that you would like to share about Matillion?
Matillion’s data transformation technology is arguably the greatest contribution to enterprise technology this year. Matillion ETL was actually named “Overall Data Tech Solution of The Year” by the Data Breakthrough Awards. We have reinvented data transformation in the cloud era – as the only provider enabling true data transformation in the cloud. Our team is quite proud to provide simplicity, scalability, speed and savings for hundreds of enterprises worldwide, and we look forward to continuing to support companies’ progression on their data journey – especially as they look to leverage data to forecast for the future. We are well-positioned to help them weather the currently turbulent global markets, and come out stronger on the other side.
Thank you for these phenomenal answers, readers who wish to learn more should visit founder of Matillion.
- Removing Objects From Video More Efficiently With Machine Learning
- AI Hardware Technology Imitates Changes in Neural Network Topology
- Stealing Machine Learning Models Through API Output
- NeRF: The Challenge of Editing the Content of Neural Radiance Fields
- Researchers Create Self-Propelled Artificial Cilia