Anyone with experience with systems thinking knows the danger of optimizing a component or subsystem independently and in isolation. In the end, while that subsystem might run more optimally, the overall system might run less efficiently or effectively.
It’s an interesting paradox, and one that requires an evaluator to take a step back and look at the overall picture, from end to end. And, when it comes to optimizing your data management strategy and its processes, it’s an absolutely critical best practice.
In my role as director of revenue operations at LeanData, I often get brought into conversations with both customers and late-stage prospects as they map out their data movement processes and the tech stacks that empower them. A frequent conversation point often involves stepping back from considerations of possible vendors and potential solutions, and starting with the overall data strategy itself.
How do you need your data to empower your organization to meet its revenue goals?
Embracing your “data swamp”
Recently I had a customer speak as to how she desired to have her marketing automation solution not be a “data swamp.” Above all, she wanted her marketing automation solution to contain only clean data, with all records containing fully completed, standardized and normalized fields.
It certainly makes sense. However, I responded back with, “Where do you want your data swamp to be?
Rather than running away from a data swamp in your system and processes, you should embrace the idea of an organized, structured and thoughtfully nurtured data swamp, and here’s why: If you insist on having only clean data in your system, you’re limiting the data you have access to and its ability to guide effective decisions.
Introducing the minimum viable data point
In thinking about data and its data journey through your processes, just because data isn’t complete doesn’t mean it doesn’t tell a story. Keep in mind, GTM operations are not Finance or Accounting, and we don’t need data to be 100 percent correct and auditable. What we need from our data is to conduct meaningful and correct analyzes.
When conducting data analysis, the minimum viable data point is the threshold level of criteria to be met in order for the data to advance to the next stage – either in your data flow, your analysis, etc. And, those incomplete data records often contain enough information to provide critical directional insights – enough to problem-solve and make decisions.
For example, you might experience a lot of drop offs on a landing page with a form fill. As the form hasn’t been completed, you don’t have a full record with name, email address and more, but you do have outstanding directional data: your landing page isn’t converting as expected. Your directional data has given your team something to problem-solve around, identify root causes, implement corrective actions, test and deploy.
Don’t be afraid of bad data (it’s just data on its journey to be good)
By bringing in incomplete data, you actually afford your team a great deal of flexibility. You just need to ensure that records meet a certain minimum viable data point before advancing through the data journey.
Wherever your “swamp” begins, use applications to manage its maturation, including normalization, enrichment, de-duplication and more. Once you’ve developed a data strategy with a data journey that supports it, you then look for the right solutions to your needs (and the right vendors to supply them).
One key factor to consider in your tech stack is your willingness to move data out of and back onto your platform. There are data security risks, of course, but also considerations around processing time and more.
Then, of course, there’s always the consideration of consolidating your tech stack. However, that’s not always the best systemic solution. Don’t use that decision variable to drive your strategy. Rather, use your strategy to drive that decision variable. Invest in data solutions (and their vendors) that are forward-thinking, scalable and work with your data movement requirements from end-to-end throughout your process.
Data isn’t always perfect when it is acquired. However, it can prove valuable, both at the time of acquisition as well as through its maturation journey. Enable your system to acquire directional data and analyze it when it’s of value to your strategy. Then, as data matures and meets minimum viable data point requirements, move its next stage in your process or analysis.
A thoughtful, end-to-end perspective of your data management flow allows your team to make data-based decisions to reach your revenue targets. And, while perfect data is always nice, it’s not always realistic. Embrace your data swamp, work and refine your data, and let it mature. The data journey should provide all the information needed for how the data got from stage to stage, and how it was used to make decisions.