The Data Science Maturity Curve is a tool developed by our partners at Bardess Group. I recently interviewed Daniel Parton, Lead Data Scientist with Bardess, for more insight about each stage of the curve.
Data Science Maturity Curve – Phase 5: Cultural
The final stage of the Data Science Maturity Curve is labeled “Cultural.” In this utopian stage, everything is seamlessly integrated: data science + business teams, data sources, applications and goals. The latest in ML and AI technologies support all goals. The company is seen as an innovator. Value is delivered, outcomes accelerated. While many might not make it to this phase, it’s an aspirational vision that can help guide programs, said Daniel Parton, lead data scientist with Bardess Group.
A major accelerator toward phase 5 is the right technology, tooling and integration, Parton explained. Some tech choices can even impact cultural goals. For example, tech can help immensely with the translation and communication between data professionals and business people. The right technologies also help make data discoveries more consumable and usable by the business, he explained.
“BI [business intelligence] tools help make visual analytics — that put data in terms which are easier to understand and more consumable. It’s also where operationalization comes in, like batch pipelines, API integrations and so on.”
The “right” architecture or platform clearly depends on specific requirements. That’s why flexible options like Cazena’s Instant Cloud Data Lake for Analytics and ML have been helpful in many different Bardess consulting engagements, Parton said. Cazena makes it easy for data science teams to use a wide variety of best-of-breed data analytics and ML tools in a secure, centralized platform.
“Having a sort of an integrated platform is so helpful, because we’ve talked about quite a few different types of technology… There’s no single sort of piece of software which does all of that, despite what some vendors might claim.
That’s where company like Cazena has done some great work — integrating these different tools into this wider platform, which is very easy to stand up for clients. So you can go from having nothing to a data stack, which just completely set up for accomplishing data science tasks.”
There are certainly alternatives, but those often require a surprising amount of integration. There’s a real “art” to integrating tools together, Parton explained.
“If you’re trying to make one tool or play nicely with another, that’s one thing. But if you have like five or six different tools or more, it can be a real pain to go through that integration to get those working together nicely. The problems take all sorts of forms — networking issues are common, security issues, single sign-on…data mapping…”
Overcoming integration challenges is critical for companies that aspire to the Cultural maturity in Phase 5. It’s a big leap, Parton acknowledged, with few in the category.
Obvious examples of Phase-5-worthy organizations, Parton said, are Google and Amazon where data science is part of the everyday language. At those companies, Parton said, most managers are well-versed in the language of data and analytics, even if they’re not practicing data scientists. They talk the talk, and consider new moves through a data lens. Having the right tech stack is also a critical differentiator, he said and seamless integration between data science and other business workflows.
“The reason those companies have been able to be so successful in data science is because it has become a kind of a cultural aspect of the companies themselves.”
Feeling Inspired? Learn more about CDAO Jumpstart. Accelerate your climb up the Data Science Maturity Curve!
Watch this Data Science Maturity Curve interview excerpt to hear Daniel share more about “Phase Five – Cultural” – and hear more about the shift from integrated, to cultural.