Collecting, validating, and analyzing any large data set can be a daunting and time-consuming task. To help focus and realize results faster, we recommend following an iterative and incremental approach. Completion of your portfolio can be extremely challenging, and chasing that number will ultimately prevent you from seeing value sooner.
There are many ways to break a portfolio into subsets:
Identify the varying criticality of components in the data set. Focus on documenting the most critical components first and working down the priority level.
Focus on one area of your business - e.g., revenue management or a different step in the customer journey - and document all the related components. Iterate through the data set until complete.
Identify one type of data attribute or relationship type on all components and focus on documenting that.
There are many approaches, but your objective is to select which approach is best. For the sake of ease, we’ll go through the first option of identifying critical components. Follow these two simple and effective steps to start turning your data into meaningful insights.
1. Establish Priority
With large data sets, we recommend identifying a criterion on your data that is important to your organization and that allows you to set its priority. An example of this could be to prioritize an application portfolio based off a Service Level Agreement (SLA) or Criticality.
Take this criterion and segment your data set into prioritized groups. This will allow you to focus on the most important parts of the data set first and will allow you to show valuable insights faster because you are not waiting on collecting and validating the entire data set.
Criticality Level | # of Applications |
Platinum | 59 Applications |
Gold | 75 Applications |
Silver | 281 Applications |
Bronze | 585 Applications |
The number of important applications should be relatively small compared to the lower criticality. This is good news because it means a focus on the Platinum applications first will see value faster.
2. Define a Success Rate
Now that you have segmented your data set into prioritized groups, it’s time to set an ideal success rate, or completion rate, for each one. You don’t need to have 100% data completeness on data you have assigned a low priority.
For example, because Platinum is the most important group of applications, you want to make sure you have the most accurate data on those applications. You don’t need that same level for your Silver or Bronze applications.
Criticality Level | # of Applications | Success rate % |
Platinum | 59 Applications | 100% |
Gold | 75 Applications | 95% |
Silver | 281 Applications | 85% |
Bronze | 585 Applications | 75% |
What this means is that your Platinum applications should have a completion rate of 100% so that all the metadata, references, and ownership for those 59 Platinum applications are documented. The Bronze applications, however, might only have a 75% completion rate but cover a higher number of applications than Platinum applications.
This does not mean that the documentation process is done once the threshold is met, but only that there's enough data to perform analysis and move on. This is a “good enough” approach that is iterative. Maybe the first time through, you only document Platinum applications, and the second time through, you look at Gold and Silver. Each pass-through shows more and more value and gives you the criteria for your entire portfolio.
This agile approach can be used on any large data set. Since it is not just limited to application data, this will help tackle any large problem by breaking it up into small pieces that we then prioritize.
Want change even faster? Reach out to us--we're happy to help answer any questions via our in-app chat or our website. 💬