Innovation Stories from the Bluemix Garage

Innovation Stories From The IBM Bluemix Garage Holly Cummins Technical Lead, European Garages @holly_cummins IBM Bluemix Garage | 2017

@holly_cummins

I’m an IBMer. @holly_cummins

I’m an IBMer. I’m a techie. @holly_cummins

I’m an IBMer. I’m a techie. I like doing things which have never been done before. @holly_cummins

I’m an IBMer. I’m a techie. The world’s most computationally advanced hat. (It can serve web pages.) I like doing things which have never been done before. @holly_cummins

I think about technology a lot … … but even I know technology should be about making things better for people. IBM Bluemix Garage | 2017 3

I’m from the Bluemix Garage. IBM Bluemix Garage | 2017 4

What is a Bluemix Garage, anyway?

The Bluemix Garage is IBM’s consultancy with a startup DNA. IBM Bluemix Garage | 2017 6

The Bluemix Garage is IBM’s consultancy with a startup DNA. (We don’t really work in a garage.) IBM Bluemix Garage | 2017 6

… but we are unique. People | Places | Platform | Practice IBM Bluemix Garage | 2017 7

People IBM Bluemix Garage | 2017 8

Designer People IBM Bluemix Garage | 2017 8

Developer Designer People IBM Bluemix Garage | 2017 8

Developer Designer Architect People IBM Bluemix Garage | 2017 8

Developer Customer Designer Architect People IBM Bluemix Garage | 2017 8

Developer Customer Designer Architect People Collaborate • across disciplines • with clients IBM Bluemix Garage | 2017 8

New York City / San Francisco / Melbourne / Nice / Austin / Toronto / London / Tokyo / Singapore Places We are embedded in startup communities. We believe in creating intentional spaces for innovation. IBM Bluemix Garage | 2017 9

Platform IBM Bluemix Garage | 2017 10

Platform Bluemix is the platform that enables us. IBM Bluemix Garage | 2017 10

Platform Bluemix is the platform that enables us. • Cognitive. IBM Bluemix Garage | 2017 10

Platform Bluemix is the platform that enables us. • Cognitive. • IoT. IBM Bluemix Garage | 2017 10

Platform Bluemix is the platform that enables us. • Cognitive. • IoT. • Analytics. IBM Bluemix Garage | 2017 10

Platform Bluemix is the platform that enables us. • Cognitive. • IoT. • Analytics. • DevOps. IBM Bluemix Garage | 2017 10

Practice IBM Bluemix Garage | 2017 11

Practice We blend three effective methodologies. IBM Bluemix Garage | 2017 11

Practice We blend three effective methodologies. • Design Thinking IBM Bluemix Garage | 2017 11

Practice We blend three effective methodologies. • Design Thinking • Lean Startup IBM Bluemix Garage | 2017 11

Practice We blend three effective methodologies. • Design Thinking • Lean Startup • eXtreme Programming IBM Bluemix Garage | 2017 11

Lean Startup.

Release early. Release often. Unreleased development effort is waste.

Minimum Viable Product. The smallest thing you can build to test your hypothesis. You do not want to invest until you are sure you are building what you know your users want and will use.

Minimum Viable Experience

IBM Bluemix Garage | 2017 Confidential

Minimum Viable Product. It’s not just the same amount of function, for less budget.

Goal

Business Opportunity We will improve the wedding cake experience by means of the icing for couples getting married. Goal

Business Factors Expensive market to enter, low volume, high cost to produce Goal

Assumption Icing qualit y influences buying decisions on wedding cake. Hypothesis Goal If we make a better icing, people will buy our wedding cakes.

Problem How do you figure out the best icing in a low volume business? Goal

Goal Minimum, but not a product Minimum Viable Product

The MVP may not share code with the final product. Experience is similar, technical underpinnings may not be.

An MVP is not a prototype. It needs to be viable. … but we can talk about ‘viable for what’?

Pre-totype Do you need to build anything at all, to test your hypothesis? Or can you just tell people about it and see what the market reaction is?

“Pay no attention to that man behind the curtain!”

Wizard of Oz

Lean Startup tells us to build the smallest thing to answer our business question. But what’s the right question?

Design Thinking.

Innovation is not achieved simply by writing an app. It has to solve a real problem for a real user.

“Designers…don’t try to search for a solution until they have determined the real problem, and even then, instead of solving that problem, they stop to consider a wide range of potential solutions. Only then will they finally converge upon their proposal. This process is called ‘design thinking.” DON NORMAN IBM Bluemix Garage

Observe Reflect Make IBM Design Thinking is a framework for delivering user-centered outcomes at speed and scale.

Observing is about immersing yourself in your users’ world. Meaningful outcomes come from an understanding of the real problems your users have. IBM Bluemix Garage

Reflecting is about converging on a point of view. Good decisions begin with an understanding of your situation, your purpose, and how you can make a difference for your users. IBM Bluemix Garage

Making is about giving form to ideas. You can’t know everything but you do know some things, so jump right in. The earlier you make, the faster you’ll learn. IBM Bluemix Garage

Manual stock checks can be very expensive.

RFID can be a great solution.

But … some stock is too squishy for RFID tags.

We can train Watson to understand our business domain.

We always start by thinking about the user. We align on one user to focus our efforts.

We generate a lot of ideas.

Then we refine and align. Everyone draws. Even the developers.

Then the Garage delivers it. This is where the collaboration between developers, designers, and architects pays off. The savings app was developed in 8 weeks, and released in 1Q 2017.

Sunaas Rehabilitation Hospital

Opportunity Patients who need rehabilitation and physical therapy forget how to do their exercises. Follow-up appointments are inconvenient.

A big retailer … with a lot of employees and systems.

All help requests were sent to a centralised email address.

Several full time staff had the job of reading the emails to route them to the right team. Creative Commons 2.0, Toms Bauģis from Galway, Ireland

This is a perfect job for machine learning.

Watson was able to classify 85% of the emails correctly. When it didn’t understand an email, it knew it didn’t know.