Toronto-headquartered global data management and governance provider Ataccama has launched the second generation of their Ataccama ONE platform. The new version extends significant capabilities which were first introduced in the first generation of the company’s modern platform, in 2017. The expanded AI and machine learning capabilities automate much that was still manual in the first platform and greatly broadens the capabilities of the unified fabric.
Ataccama is a subsidiary of large IT service management provider Adastra, which developed the original data management platform and spun it out into a subsidiary company in 2007.
“It was me on both sides, creating a spin off and being nervous about it because I was head of sales for Adastra, who was taking my 10 best people out and taking risks with a new company,” said Michal Klaus, who became and remains Ataccama’s CEO.
“We really started in 2008, and the market fall played to our advantage,” he said. “Back then, when times were good, companies just went to IBM and paid them what they asked for. When times are bad, they have to take a closer look at what they are paying.”
Klaus said that the original concept that Ataccama would build the product and Adastra would do the sales and marketing didn’t really work, and so they built those out themselves. They also built out the capabilities of the platform, starting with Data Quality Management in 2010. They got a big boost in 2011, when they landed a “major, major customer in Europe” – Lloyds Banking Group. That came just after landing the Bank of Montreal.
“Our focus is on medium to large enterprises on a global scale,” Klaus said. “We work best for companies with a lot of data and where the data is critical for them.
Most of their revenue – at least 80% comes from direct sales. Their channel partners are a combination of resellers and integrators. OEMs – Information Builders and Infor – were once significant but have declined as a percentage of the whole and are now the smallest revenue component.
“A big differentiator for us was always that we had a single engine organically built in-house that did data cleansing and data management,” Klaus said. “Most of our competitors had two separate products. Informatica, our biggest competitor, still has a separate DQM and a MDM, and someone has to integrate them.”
Klaus said that a major milestone for the company came at the end of 2017, when they introduced the original version of the modern Ataccama ONE platform.
“The big change at that time was that we introduced a new layer which tied all the modules together,” he indicated. “The data stewardship modules were previously separate applications. This joined them all together on the front end. Back then, we also decided we would give access to the metadata in the platform to the data stewards, which was our first data catalogue.”
Now, with the second generation of the ONE platform, Ataccama is extending what it did in the first generation.
“Compared to the previous integration where we integrated modules, we have rethought the processes completely,” Klaus said. “We decided to change the concept from self service application for data stewards because we wanted to apply AI more. So we went from self-service to self-driving. The concept here is that AI in the platform will figure out the steps that it needs to take. Our ambition is that you tell the platform what to do and it does it.”
Klaus stated that this enhancement of AI and machine learning capabilities with the new self-driving data management now provides a differentiation from their competitors.
“Before this, our AI was comparable to our competitors, in that it was the point application of AI in the portfolio,” he said. “With Generation 2, we completely revamped the concept. Having the AI drive the whole thing is the big difference.”
Klaus also emphasized that the level of integration also goes well beyond the first version, which just integrated all the modules together. The unified fabric in this one integrates all the disparate disciplines of data governance and data management, including data quality, master data management, data catalog, data governance, and data processing.
“What we had before was a data processing engine we used for data profiling, and it was all done with one engine,” Klaus indicated. “Now it is at a different level because we have the modules, and alongside these a brand new metadata management and data governance capability in the form of policies. It’s the same principle, but at a completely different level. The platform will put together the processes and give you the result, if you have defined the policies.”
Klaus also indicated that the platform now effectively has no boundaries for data because it processes the data where the data lives
“We have rewritten our engine in such a way that it has a dispatcher to deploy tasks to whatever processing nodes it has available, whether it’s an in-house data lake, or a Teradata warehouse,” he said. “The engine will be aware of what data is stored where and will process it where it makes sense. We know of only one other competitor, and it’s a smaller one, which processes the data where the data lives.”
Ataccama is now offering a limited beta program for the new Ataccama ONE platform. Selected participants will receive six months for free. General availability is planned for February 2021.