Nicolette Celli, Ultimate Software, USA
Patrick Alt, Ultimate Software, USA
In any business, understanding the health and quality of products and teams is essential to success. However, when we have a large number of teams working independently, we run the risk of duplicating data or having too many sources of information, such as automated test result logs and documented test inventories. Teams use different technology stacks, architectures, and pipelines, and these require unique solutions for evaluating information and gathering metrics. Without a centralized source of information, this often leads teams to independently find ways to attain the data they need. This poster presents how we approach the complex “quality puzzle” and organize our services to avoid confusion, inefficiency, and wasted time and effort. With this approach, the services rely on a source of truth to pull data from, where all data is accurate and up-to-date. Certain standards are enforced, such as naming conventions, in order to unify information across all teams. We use Elasticsearch to dynamically aggregate all data on product, team, and domain levels to display on a company-wide quality dashboard. In the future, we will be able to easily determine whether a team is healthy quality-wise by using machine learning to automatically cluster related data.
Learning:
- Determining the health of a team, product or domain
- Organizing and unifying services
- Importance of standardization in large companies
- Making metrics easily accessible for everyone
Nicolette Celli, Patrick Alt, 2019 Poster Paper
