Enterprise Awareness

Background

I identified a lack of operational intelligence and real-time awareness in our day to day operations while leading outages. While we could look up individual user sessions easily, the process was slow and didn't scale at all. Finding out how many people were logging in and working in a massive, international call-center environment is really important and timely data is critical. Being able to rapidly analyze historical data and look for clusters or quantify impacts was an awesome aspiration for us.

The Project

After proposing a high-level solution to replace our inadequate existing solution and gaining executive buy-in, I validated the technology with a proof of concept. I compared and contrasted options using SQL, NoSql and Application Insights from Microsoft (the Kusto query language now.) I also focused on expanding the collected data from only a heartbeat and 3 dimensions of data to include multiple metrics and a much broader selection of dimensions that were hierarchically embedded, providing much richer data without dramatically increasing the storage footprint.

I was privileged to lead development efforts as the Product Manager for the team - I was able to ensure that user needs were kept as the impetus of all efforts and when we made technical trade offs, cost, capability and experience were balanced. While I had initially wanted to code a lot of the platform myself, a wise manager pointed out that while I can code, I'm not as efficient as the engineers - but I'm much stronger with the vision and people fronts.

The Results

The project was a total success: while realizing massive improvements in efficiency as storage demand fell by 37%, query speeds increased significantly thanks to normalization, and we were able to retain permanent high-level session data for the first time ever.

While I coded almost none of the production system, leading the project using only influence and getting to set the roadmap and plan features was an amazing experience. I learned that I can have a bigger impact by working to my strong points than by only working on the actual coding and that a great team can deliver much faster and more confidently. This is the project where the product bug bit me.

Having a strong technical background and the respect of the technical team meant I was able to keep up and make informed decisions with the engineers but being able to focus on the roadmap and long term outcomes meant that I could make sure we got where we needed to go. We continued to add features as resources were available and needs were identified. This system is still in used every day by dozens of key support people two years later and is a core part of our enterprise awareness and reporting tools.