The rise of Data Product Management
Why do you need a data product manager in your organization?
There are two types of companies in the world. The ones that follow the principles of product thinking and the ones who don’t. The same logic same can be applied to people.
Personally, I don’t spend too much time explaining and reading about what product management is and why is it important. Instead, I prefer to see it as a way to structure a roadmap, adapt while progressing, and ensure the organization is obsessed with the user’s needs and day-to-day problems.
What is a data product manager?
Data product managers (DPM) are responsible for identifying gaps in their internal user’s data experience and bridging them by working with the machine learning and data engineering team.
Yet, they will differ from the traditional product manager in the same way data products differ from traditional software, specifically on these points:
Stakeholder’s expectation management.
Data products are built through experimentation processes. On the data engineering side, you’ll never know the complexity of the pipeline you will have to build. On the machine learning side, you’ll never know if the collected data will enable a successful model. So, there’s no other way than test it and see the results.
This means the way expectations are managed outside the data team, it’s fundamental.
The required technical expertise
For a product manager, it’s important to, first, be able to communicate in the same language as the technical team. Second, be able to understand the feasibility of a solution.
When applied to data this means that the DPM should understand topics such as data modeling, the different sets of services needed to establish a data architecture, understanding how a model is created, and how all of this will be deployed into production. Practical speaking, the DPM should be comfortable with python, SQL, and modeling, so that product feasibility can be tested before it comes to the roadmap and when the product needs validation before production.
It’s also important to highlight those skills are also essential so that the DPM can reply directly to business questions with a high level of confidence.
Senior and Middle management education
If you are in the data and machine learning business then you are in the education business. Not everything around data will be data stacks or data/machine learning pipelines.
Enabling your organization and users is key to ensuring there are fewer obstacles to adopting any data product. This is usually the reason behind the continuous and eternal excel usage. (I love excel by the way)
Focus on Data Trust
Ensuring a single source of truth has never been so important. In the last decades, data flowed from operational systems, IoT and others applications to staging areas. The famous ETL or ELT. Then it was moved to data warehouses where there was zero ownership over the tables with information.
When something wasn’t correct, the machine learning engineers were fixing the problem in the machine learning pipelines rather that trying to solve the problem at the source. This resulted in huge data silos, different sources of truth, unreliable data products, and a huge data debt.
Therefore, liaising with other data product managers and different data groups within the bigger data team is key to ensuring a clear line of cooperation.
How data products and data product management will shift the future of data in organizations?
Below you can see the most relevant benefits of adopting product thinking to the way data products are developed.
The future of data teams
The distance between data producers and data users is growing and demand is increasing exponentially. This is due in part to the growing reliance on data across all parts of an organization.
Data teams should become a conductor that spans silos and the data product manager should inspire teams to play in harmony.
As a result, organizations will move from a reactive posture of fighting data fires to a proactive stance of building internal data capabilities as a competitive advantage.
If you want to read more content make sure to follow me on LinkedIn for more weekly posts, and if you liked my article, please consider subscribing. I post once a week, every week.
Thank you so much for reading and let’s talk again soon.