The Customer As System Under Test

30-minute Talk

All levels

Timetable

11:10 a.m.-11:40 a.m.

Tuesday 5th

Room

Room F2 - Track 2: Talks

Audience

Tester, Product Owner, Product Manager, Developer, Manager

Key-Learning

- How to gain trust and buy-in from organizations that are used to disconnecting strategy from execution and disconnecting the deciders from the doers.

- How to change the accountability to the impact on behavior the deliverables will have rather than accountability to the deliverables themselves.

- How to get people thinking in terms of hypotheses, bets, and experiments rather than just deliverables, dates, and the cost/time/scope triangle.

- How to involve the customer from the earliest stages of building your product, when you first begin to consider potential strategies.

- How to create small experiments that limit what you have to build to test an assumption and let you keep decisions easily reversible.

The Role of the Tester In A Lean Startup/Enterprise

In a Lean Startup or a Lean Enterprise, validation becomes just as iterative as verification. This expands the role of the tester to someone who can help bring the testing mindset to the whole team.

In a Lean Startup or a Lean Enterprise, validation (did we build the right thing?) becomes just as iterative as verification (did we build the thing right?). This greatly expands the role of the tester to someone who can help bring the testing mindset to the whole team as they prototype, experiment, and decide if and when to pivot, all of which requires the team to be able to connect strategy to execution.

In this talk, I will connect strategy to execution using two kinds of mapping: 1) Gojko Adzic’s Impact Maps and 2) Jeff Patton’s Story Maps.

This will be an experience report of applying the mapping techniques in a large enterprise (Salesforce.com) working in a space with high volatility, uncertainty, complexity, and ambiguity, specifically, building an entirely new data and compute pipeline to do machine learning at scale on a just-in-time stream of production and performance testing data.