DataOps predictions for 2025 - N°1: The Fog Clears, But Slowly

Oh yes, it’s that time of the year ;) In the weeks to come we will publish some predictions for 2025. Number 1: The Fog Clears, But Slowly.

The most important reason we made the Industrial Data Platform Capability Map a few weeks ago was to put a stake in the ground and help ourselves and you to start a meaningful data platform discussion.

Industrial Data Platform Capability Map (v1)

2024 was definitely the year where everybody started talking about Data Platforms and DataOps. The year where the “traditional” solutions started getting some serious heat from start-ups, scale-ups and even the hyper scalers.

This has turned the industrial data domain into a murky landscape, filled with sweeping claims from all directions:

• “Our solution can do everything you need”,

• “We have adapted our existing solution to now also work with sensor data”,

• “We only use open-source components”,

• “We fully adhere to the Unified Namespace approach”, etc, etc, etc…

It’s fun to write this blog as it gives us very good insights into what vendors are working on and how they are presenting themselves. Because this topic is so important right now, we will focus more on DataOps in 2025:

• We will keep on extending the Capability Map (first step is to extend towards the data science/visualization/sharing stuff).

• We will start building a database with vendors who are active in this space, and we will also start recording interviews to give them the possibility to present themselves and their capabilities 📹.

• We are currently creating a one-day training, this will become general availability somewhere mid 2025 and are starting to do some try-outs right now (if your company is interested in booking such a try-out, contact David, only a few spots available).

Willem doing what he likes to do in Barcelona: using flip charts and post-its (and trying out one of our talks about Data Governance and Unified Namespace ;))

🎉 You can now use and adapt our map!
To help with the adoption process of the Capability Map, we have now intentionally labelled it as CC BY-SA 4.0 which means that everyone can freely use, share and adapt it as long as you attribute us, and you distribute your work again under the same license.

Our Industrial DataOps Predictions for 2025 (text below video):

• Progress from "Dark Grey" to "Grey"
  While selecting a comprehensive DataOps solution will remain challenging, we anticipate significant improvement. End-users will continue to grapple with identifying their core requirements especially with all the noise to be found on the internet, assessing internal capabilities, and choosing the right partners (because they can still “do it all” 😉). However, with clear focus and informed decision-making, overcoming these hurdles is becoming increasingly feasible.

• Maturing Scale-Ups Take Center Stage
  We foresee multiple mature scale-ups offering solutions that address most DataOps capabilities from our map. These companies will drive innovation, providing more cohesive and integrated offerings, making it easier for organizations to deploy effective DataOps strategies.

• Tough Times for Traditional OT Vendors
  Legacy OT vendors are likely to face a challenging year as they struggle to defend their market positions. Increasing scrutiny will spotlight gaps in both their technical capabilities and pricing structures, pushing them to adapt or risk losing relevance in a rapidly evolving landscape. These vendors will need to open up to community & client input, partner up with strategically chosen start-ups and scale-ups and be ready to play a role in this rapidly evolving landscape.

• Cloud Hyperscalers Expand Their Reach, But Caution is Key
  Cloud hyperscalers will continue to assert their presence in the Industrial DataOps domain, leveraging their vast infrastructure and capabilities. However, it is crucial to recognize the distinction between simply offering data storage and compute resources and genuinely understanding the unique complexities of the manufacturing and process industries.

• Data Engineering and Data Science Enter OT (but not because of the AI hype)
  Data engineering and data science are set to become staples in the OT domain, with organizations actively dismantling barriers between IT and OT data. In many cases driven by the need to become a more competitive player, especially in the current economic conditions. We’ll see new teams emerge, blending skills and insights from both worlds to unlock the full potential of industrial data. However, this shift must be handled with care—collaboration, not domination, is the key. It’s essential to create a balanced approach where neither IT nor OT overpowers the other, ensuring effective teamwork and shared ownership of outcomes.

We’ll be back soon with our next prediction ! Stay Tuned :)