It should come as no surprise that most companies are using Agile methodologies in some capacity these days and it is quite clear why.
- At least 71% of U.S. companies are now using Agile.
- The success rate of Agile projects is at 42% compared to 14% for traditional Waterfall projects.
- Teams that use Agile are 25% more productive on average.
Even though agile is highly adopted and successful, organizations are still working on the best way to manage data initiatives. Organizing your data projects into data products & portfolios is simply the first step. From there, it is important to respect the nuances of working with data.
Fly the plane while building it
First, two questions.
1. Do you trust your data?
Most organizations will answer “no” or “sometimes.” As a result, many organizations try to make all their data high-quality before even attempting to deliver value. I liken this to trying to boil the ocean.
2. Do you have the data products you need to do your job to the best of your ability?
Most people will say “no.” As a result, many organizations focus only on delivering user-facing data experiences leaving data consistency and scalability on the back burner. This is like building a house without a foundation.
When I say fly the plane while building it, I mean treating data as a product requires that organizations must build data foundations one data product release at a time. Yes, this can be more inefficient than building all your data foundation at once. Yes, this will make your product releases take longer. However, the alternatives can result in lost trust, lost investment and lost time.
Mature one step at a time
When I was a kid, I used to try to go up two or three steps at a time for some odd reason. Fine, I was trying to be cool. AI is cool (just like I was) and organizations want to get to using it as quickly as possible. Unlike me on the stairs, skipping steps along the Data & AI Maturity Scale is not cool.
AI requires very good quality data and a ton of business context to add value and do no harm. Climbing the maturity scale one step at a time is the best way to deliver iterative value and build your data foundations to enable AI applications. Below is our Data & AI Maturity Scale.
1. Proof of value
This level is about determining whether your product will deliver value and how. It’s important to identify the business problem you’re trying to solve, the metrics you’ll use to measure success and confirm that with the users before ‘building’ anything.
2. Curated insights
At this level, insights are delivered to the business from analytics professionals. The focus is on creating a culture of data-driven decision-making and improving the quality & understanding of the data through usage and adoption.
3. Self-service analytics
Defined and trusted data is made available automatically for end-user consumption. This level is about empowering business users to evolve data-driven decisions and understanding the limitations of human-driven analysis.
4. Data science and machine learning
At this level, complex analysis is done using machine learning algorithms and data science approaches. The focus is on making decisions better fueled with better answers via data.
5. AI & Automation
At this level, enough business context exists alongside trusted data to enable safe and valuable AI & Automation. The focus is on developing intelligent systems that can automate business processes and augment human decision-making.
Achieving value with Data & AI is very complex and that is why we started Origin. We believe when you treat data as a product, you can succeed and avoid common setbacks. For help instilling data as a product mindset at your organization, contact us at email@example.com.