Reading “Your Data Supply Chains Are Probably a Mess. Here’s How to Fix Them.”

A good software testing strategy is like a supply chain data pipeline.

The purpose should be to get the relevant data in the right hands so they can make decisions about quality and risk.

Here are the common challenges:

1. The actual technical process of developing automation can be overwhelming and you can lose sight of the big picture in the implementation detains.

2. The right data to perform meaningful tests is often locked away in different silos. Whether developer knowledge of APIs or business understanding of requirements.

3. A common data communication language is necessary to communicate business priorities to QA and for QA to communicate their findings in a way that can provide contextual meaning to various stakeholders.

4. Different parts of the organization have different goals, and aligning them all with meeting customers need for quality and business need for the bottom line.

Solutions:

1. Leadership should drive the need for QA by communicating thier priorities and assigning test resources according to business goals and value.

2. QA should be demand driven. Developers, product owners, and leadership should seek the knowledge they need to make decisions from testers and enable them with the information needed to accomplish the testing.

3. Testers and developers should understand the business domain language and common communication channels should be open. If test reports and continuous delivery jobs are ignored, find out why: is the data accurate, relevant, timely, and meaningful?

4. QA leadership should align testing with customer and business leads and approach testing as a source of information for product decision makers not as a gateway or “check” on software quality.

The data you need to perform tests and the data you need to make decisions about quality are inter-related but not identical.

The ability to share (and transform) data for testing is critical, but I don’t think a unified tool or process is the solution. It’s why complex ERP deployments fail and why everyone hates Jira.