Agile at scale May 11th 2016

Tomas

Agenda 18:05 10 min

Welcome!

18:15 35 min

Scrum@Scale / Spotify, by Tomas Björkholm

10 min

19:00 35 min 10 min

19:45 35 min 10 min

Q&A Scrum@Scale

SAFe (Scaled Agile Framework), by Jan Grape Q&A SAFe

LeSS (Large Scale Scrum), by Michael Göthe Q&A LeSS

20:30 10 min

Certified LeSS Practitioner, June 15-17

20:40 40 min

Discussion areas: Scrum@Scale, SAFe and LeSS

Jan Grape

Janne

Michael

LeSS with Craig Larman

Jan Grape

Upcoming book § Authors Craig Larman & Bas Vodde

§ Pre-order at Amazon § LeSS practitioners get access to draft version now

Upcoming class - Certified LeSS Practitioner Principles to practices

§ June 15-17 here at Crisp § With co-creator Craig Larman § In-depth about scaled agile from one of the foremost experts § LeSS principles, frameworks, guides, experiments, practices, rules, …. § Sign up at https://crisp.se/kurser

3 frameworks

3 Scaled Agile frameworks

Scrum@Scale

Tomas Björkholm

House of Agile with good precision and low risk New knowledge Mobilizing brainpower

Fast feedback

Be able to react on new knowledge

Continuous improvement (adaptation) Collaboration - High team autonomy

Tomas Björkholm

Collaboration vs. Autonomy Example: How to build a road over the river We’re using the same tools, sprint length, board structure, team size, team setup … - all to make collaboration easy.

Let’s build a road and a bridge to get over the river

Let’s help each other get over the river. Some standards are needed.

High Collaboration

We don’t know what the others are doing, we’re just using our standard process and set-up, to build this road.

Maybe some other team is also trying to get over the water somewhere else. But we don’t care.

Low Collaboration Low Autonomy

High Autonomy

Tomas Björkholm

LeSS a nutshell Michael Göthe

Background

Michael Göthe

LeSS Core concepts § Inspect and adopt framework for organizations § Stay true to Agile Manifesto and Scrum § “Barely sufficient methodology” § Product is central

Michael Göthe

Overview

Michael Göthe

Principles

Michael Göthe

Principles

• Multiple-team Scrum, not multiple Scrum teams • Scrum principles applied to org/product

Michael Göthe

Principles

§ Less roles § Less artifacts § Less process § Less bureaucracy § Less complexity § Less organizational debt

Michael Göthe

LeSS Framework

Michael Göthe

LeSS Huge Framework

Michael Göthe

Structures

Michael Göthe

Structures Specialization in the customer domain to increase collaboration with real users and remove hand-offs • E.g One feature team handling all platforms

Michael Göthe

Technical excellence

Michael Göthe

Technical excellence CI - Working system •

by reducing variability, uncertainty and risk on product level.

Michael Göthe

Management

Michael Göthe

Management Grow organizational capabilities •

Does not manage teams

Michael Göthe

Adoption

Michael Göthe

Adoption • • •

deep and narrow over broad and shallow top-down and bottom-up use volunteering

Michael Göthe

Summary § 11 Principles § 2 Frameworks § 28 +13 Rules § Guides § 600 Experiments

less.works Michael Göthe

Scrum@Scale May 11th 2016

Scrum@Scale

Tomas Björkholm

This is Scrum@Scale • 1 more meeting/group: Executive Action Team • No new roles • 9 areas/modules with goals, inputs and outputs Your work is to discuss how to, given the inputs, reach the modules’ goals and deliver expected output

Tomas Björkholm

Tomas Björkholm

Tomas Björkholm

Tomas Björkholm

Disclaimer This is a simplified version of the framework Scrum@Scale. The examples are my observations and interpretations and is not representing the full truth.

Tomas Björkholm

Tomas Björkholm

0. Executive Action Team Module Goals: • Align the development of the entire organization • Manage the high level transformation process (organize transformation backlog) • Remove impediments outside the teams’ area of influence • Support POs and SMs through mentoring, coaching, challenging, and evolve the organization

Vision & goals regarding organizational culture, structure, values, norms etc.

Business strategy and market trends

Organizational transformation strategy

Identified impediments

0. Executive Action Team Up-to-date Executive Action Backlog

Tomas Björkholm

Feedback on interventions

Examples from Spotify

Tomas Björkholm

Alliance Tribe

Tribe PO

PO

PO

PO

PO

Chapter

Chapter

Chapter

Chapter

Squad Squad Squad Squad

Tomas Björkholm

PO

PO

PO

Squad Squad Squad Squad

Tribe PO

PO

PO

POCLAC

PO

Chapter

Product owner Chapter

PO

Squad Squad

Squad Squad

What Agile coach

Chapter leads

How

Way of working Squad members

Tomas Björkholm

Module goals: • Maximize the flow of complete and quality tested work • Try to increase velocity every sprint • Work in a sustainable way

Potentially shippable increment Artifact transparency

Feedback on product increment

Team-Level Process Ordered product backlog

Tomas Björkholm

Tomas Björkholm

Module goals: • Clearly align the entire organization • Compelling articulate why the organization exists • Describe what the organization will and won’t do

Hypotheses on market needs Clear goals and principles for backlog ordering

Consumer and market insight

Strategic Vision Feedback on released progress

Tomas Björkholm

Tomas Björkholm

Tomas Björkholm

Tomas Björkholm

Module goals: • Identify a clear ordering for products, features and services

Potentially shippable increment Artifact transparency

Feedback on released product

Backlog Prioritization Current product backlog

Tomas Björkholm

Tomas Björkholm

Squad prioritization strategy 100%

Prio 3 - Squad’s own decision

Prio 2 - Bets

0%

Prio 1 - Keep The Lights On

Bets helps us do initiatives where Tribe prioritization is needed. It can either be situations where collaboration between squads are needed or as a way to protect squad initiatives that else might be down prioritized. Tomas Björkholm

Module goals (examples): • Break complex projects into manageable elements • Capture emerging requirements • Ensure backlog items are in a “ready state”

Consolidated and individual team level product backlog

Backlog Decomposition & Refinement Tomas Björkholm

Emerging requirements from feedback and brainstorming Current product backlog

Bet intake w. stake holders (four weeks before new quarter) Bet prioritization w. stake holders (one week before new quarter) Tap planning (1:st day in month)

Planning Follow up

Stake holder update (end of each month) P&A (bi weekly – Thursdays at 13:00)

March Tap

April Tap

May Tap

June Tap

July Tap

August Tap

Tap planning

Tomas Björkholm

Module goals: • Forecast delivery • Communicate snapshot of delivery expectations

Velocity for all teams

Road map Burn down chart

Release planning Ordered product backlog

Tomas Björkholm

Release planning – fixed scope Release burndown chart 400 Work remaining (story points)

300

When will all of this be done?

We’ll be done around sprint 14-16

PO

200

100

1

2

3

4

5

6

7

8

9

10

Sprint

Tomas Björkholm

11

12

13 14

15 16

User story map

Tomas Björkholm

Tomas Björkholm

Tomas Björkholm

34

Module goals: • Deliver a consistent flow valuable finished products • Integrate the work of different teams • Capture and communicate feedback on product release

Product delivery to customers Update to release plan

Feedback on product increment

Release management Steady flow of increments from teams

Tomas Björkholm

Operations

Dev Dev

Tomas Björkholm

Operations squad

Dev squad Dev squad Dev squad

Tomas Björkholm

Module goals: • Understand how users use the product • Capture ideas for new features

Result of experiments Ideas for new features

Product & Release Feedback Tomas Björkholm

Identified integration and release issues Feedback from actual users

Tomas Björkholm

But will the rest of the world like it? Dunno. We’ll find out.

1%

Tomas Björkholm

Watching the data

1500 survey responses

User metrics

Ratin g

Tomas Björkholm

Watching the buzz

Tomas Björkholm

Gradually roll it out again

Tomas Björkholm

Module goals: • Identify impediments that slow teams down • Identify opportunities to get faster

Visibility of impediment status Revealed learning on process experiments

Continuous improvement

Updates on impediment removals Impediments raised by individual Scrum teams

Tomas Björkholm

Module goals: • Coordinate similar processes • Manage dependencies • Alignment of team norms and guidelines

Visibility on team norms Actions to sync backlogs

Identified cross-team dependencies

Cross-Team Coordination Request for changes to norms

Tomas Björkholm

P&A board

Tomas Björkholm

9:05 meeting

Tomas Björkholm

WRAT

Sprint goal pitch line

Impediments

Anatomy

Release burndown

Aggregated Scrum board Tomas Björkholm

External Trouble Reports

Management taking care of impediments

49

Tomas Björkholm

49

Module goals: • Provide all decision makers with data to make a good decision • Shorten feedback cycles • Reach this with minimize additional effort by teams, stakeholders or leadership

Insight for updating strategic vision Context for making frontline decisions

Customer satisfaction

Metrics & Transparency Velocity data for all teams

Tomas Björkholm

Tomas Björkholm

THANK YOU!

Questions?

Tomas Björkholm

SAFe in a nutshell Jan Grape

Jan Grape

This doesn’t scale!

WTF!

#@%&?!

Hey!!

Jan Grape

If we align on a couple of things.. Same release train shares -------ü Rythm ü Priorities ü Ownership It’s easy to coordinate within a release train

We can predict when we need to coordinate with other release trains

Jan Grape

Identifying value streams Car dealerships Identify scenarios where exchange of value happens

Exposure of cars to sell

Banks

$

Every point where exchange of value happens is the beginning and end of a value stream for your company

Consumers

Potential clients

Cars to buy

Car ad site Exposure to partners

$

Assume a black box view of your company

Money isn’t always involved

Potential clients

$ Identify external parties your company deals with

Insurance companies

Jan Grape

Value stream: new customer Car dealership

Exposure of cars to sell

$

Car ad site The value stream may have started earlier with marketing etc.

Receive call

Hand-off

Sell a subscription

Register new customer

Customer service Value adding time

Credit & background check

We get our value

Customer gets his value

Activate new account

Finance

Send account details to customer

Customer service Waiting time

Jan Grape

Login to admin

Add car details

Publish!

Customer/Car dealer

Send invoice

Pay invoice

Finance

Customer

Value stream: add new product feature Car dealership

We need to optimize the whole value chain!

New way of presenting cars

$

We deliver slow! The developers have to work faster!

Car ad site

Receive feature request

Customer service

Receive feature request

Customer service

Screen feature request

Create business case

Product management

Screen feature request

Funding decision

Exec LT

Create business case

Funding decision

Project creation

PMO

Project creation

Requiremen ts analysis

BA

Requiremen ts analysis

Feasibility study

Go decision

R&D

Exec LT

Feasibility study

Go decision

Product design

Implementa tion

Testing

R&D

Product design

Implementa tion

Release creation

RM

Testing

Agile Release Train For creating new features, we form a unit along the value stream called the Agile Release Train

Release creation

Deployment

Sales

OPS

Sales

Deployment

Sales

Sales

Invoice

Finance

Invoice

Finance

Jan Grape

SAFe structure • Funding of release trains • Strategy

Strategic

Portfolio

• Cross-team facilitation • Common resources

Agile Release Train Program

Tactical

Teams

Operational

Jan Grape

• Multiple teams working in parallel • Cross functional teams • Scrum/XP

SAFe – planning cadences § Product management with three planning horizons

Kanban Company specific cadence

Portfolio Strategic

Program Increment 10-12 weeks

Program Tactical

Jan Grape

Sprint 2 weeks

Teams

Operational

§ Synchronized planning and releases for 5-20 teams

RTE

§ Potentially shippable every 2 weeks

Arch

§ Plan on a cadence, release on demand

Ux Ux

PM

Kanban

System Team

SAFe – Agile release train

PM PM

Program Increment Jan Grape 10-12 weeks

PO

SM

PO

SM

PO

SM

PO

SM

PO

SM

Sprint 2 weeks

Plan on a cadence – release on demand

2016

2017 IP

PI 1

IP

PI 2

IP

PI 3

§ Plan regularly § Every two weeks there is a release candidate

IP

PI 4

IP

PI 5

IP

PI 6

IP

PI 7

§ The IP (innovation & planning) sprint – 20% § Hack weeks § Overflow buffer - predictability

§ Release to market when business demands

§ Planning for next Program Increment Jan Grape

Alignment & problem solving Sprint 1

Sprint 2

Sprint 3

Sprint 4

Team Awesome Team Fugees Team black ops Team Knights who say ’ni’ Team Greenhorns Team Sharks

We can now resolve the dependencies required to ship

I can see when stuff is intended to ship!

Jan Grape

IP

SAFe – Backlog flow Enables agility

Strategic themes

Stakeholder input

Portfolio backlog

User Story

Enabler

Alignment across agile release trains

Enabler

Feature

Alignment across development teams

User Story Enabler

Enabler Enables agility

Feature

Business Epic

Kanban

User Story

Feature

Enabler Epic

Feedback

Team backlogs User Story

Feature

Business Epic Business Epic

Program backlog(s)

Stakeholder input

Enables agility

Feedback

JanProgram Grape Increment 10-12 weeks

Feedback

Sprint 2 weeks

Jan Grape

Classic projects § Funding of staff (capacity) is bundled with content

Project Staff (capacity)

Content

§ Money (funding) is used to decide what to build § New content requires new funding and new staff (capacity)

New project New staff (capacity)

Jan Grape

New content

Capacity vs product content Backlog

+

+





Capacity

Product content

§ The number of development teams

§ What the development teams implement

§ Slow to adjust (firing/hiring + training)

§ Anything not started yet can be replaced at minimal cost

§ Takes several months

§ Small and frequent releases § Takes weeks or even only days Jan Grape

Budgeting with stable agile teams 140000 120000 100000 80000 60000 40000 20000 0

§ Team cost is flat, roughly € 100 000/month § Budget done :-) Development §cost Value of

implemented Business value backlog must exceed cost

§ ROI is determined by value

Jan Grape

SAFe lean agile budgeting Funding of delivery capacity for each Agile Release Train Exactly what to build is decided using backlogs Private banking ART

Portfolio

Program

$$$$$$ Teams

Private banking

B2B ART Program

$$$$$ Retail banking

B2B

Teams

Retail banking ART Program

$$$$$$$$$$ Teams

SAFe components for the die hard agilist Where is the project office?

Decouple financing from content Prioritize using WSJF

Self organized cross team planning

Innovation sprint

Agile at scale 2016-05-11.pdf

Sign up at https://crisp.se/kurser. Principles to practices. Page 5 of 95. Main menu. Displaying Agile at scale 2016-05-11.pdf. Page 1 of 95 Page 2 of 95.

49MB Sizes 1 Downloads 130 Views

Recommend Documents

Agile at scale 2016-05-11.pdf
Whoops! There was a problem loading more pages. Agile at scale 2016-05-11.pdf. Agile at scale 2016-05-11.pdf. Open. Extract. Open with. Sign In. Details.

Agile at scale 2016-05-11.pdf
Download. Connect more apps... Try one of the apps below to open or edit this item. Agile at scale 2016-05-11.pdf. Agile at scale 2016-05-11.pdf. Open. Extract.

ParaView - Data Science at Scale
scientists to visualize and analysis extremely large data sets. The tool ..... For advanced users who wish to create complex program graphs, the program graph.

Dynamic iSCSI at Scale- Remote paging at ... - Research at Google
Pushes new target lists to initiator to allow dynamic target instances ... Service time: Dynamic recalculation based on throughput. 9 ... Locally-fetched package distribution at scale pt 1 .... No good for multitarget load balancing ... things for fr

Shasta: Interactive Reporting At Scale - Research at Google
online queries must go all the way from primary storage to user- facing views, resulting in .... tions, a user changing a single cell in a sorted UI table can induce subtle changes to .... LANGUAGE. As described in Section 3, Shasta uses a language c

Yedalog: Exploring Knowledge at Scale - Semantic Scholar
neck when analyzing large repositories of data. We introduce Yedalog, a declarative programming language that allows programmers to mix data-parallel ...

Software Defined Networking at Scale - Research at Google
Google Confidential and Proprietary. Google's Global CDN. Page 7. Google Confidential and Proprietary. B4: Software Defined inter-Datacenter WAN. Page 8 ...

Web-scale Image Annotation - Research at Google
models to explain the co-occurence relationship between image features and ... co-occurrence relationship between the two modalities. ..... screen*frontal apple.

Optimizing Google's Warehouse Scale ... - Research at Google
various corunning applications on a server, non-uniform memory accesses (NUMA) .... In addition, Gmail backend server jobs are not run in dedicated clusters.

Google's Green Computing: Efficiency at Scale
centers hosting cloud services are more efficient than the in-house office servers they ... The network: Wireless routers, network switches and all the intervening ... the server level energy usage.1 This paper examines server energy only.

Fleet management at scale Services
Google's employees are spread across the globe, and with job functions ranging from software engineers to financial analysts, they require a broad spectrum of technology to get their jobs done. As a result, we manage a fleet of nearly a quarter-milli

Tera-scale deep learning - Research at Google
The Trend of BigData .... Scaling up Deep Learning. Real data. Deep learning data ... Le, et al., Building high-‐level features using large-‐scale unsupervised ...

Market Reconstruction 2.0: Visualization at Scale - FIS
24 Mar 11:52:33 | |- id: string. |. |. | |- Parent: ..... at http://frozeman.de/blog/2013/08/why-is-svg-so-slow/ ... Market Reconstruction 2.0: Visualization at Scale 24.

Experimenting At Scale With Google Chrome's ... - Research at Google
users' interactions with websites are at risk. Our goal in this ... sites where warnings will appear. The most .... up dialog together account for between 12 and 20 points (i.e., ... tions (e.g., leaking others' social media posts), this may not occu

Fleet management at scale Services
How Google manages a quarter million computers securely and efficiently .... control systems. Admins can see a complete revision history of the imaging ... latest in consumer enterprise hardware, working with outside vendors and partners, as well as

Large-scale speaker identification - Research at Google
promises excellent scalability for large-scale data. 2. BACKGROUND. 2.1. Speaker identification with i-vectors. Robustly recognizing a speaker in spite of large ...

100GbE and Beyond for Warehouse Scale ... - Research at Google
from desktops to large internet services, computing platforms ... racks and clusters interconnected by massive networking ... five years for WSC interconnects.