More Business Value Now - Triad

© Joseph Little 2013

MORE BUSINESS VALUE NOW“Transforming the World of Work”

1

© Joe Little 2013

Joe Little, CST & MBA

• Agile Coach & Trainer (CST)• 20+ years in senior level consulting to well-known firms in New York, London and

Charlotte, and elsewhere.• Focus on delivery of Business Value; interest in Lean • CST (CSP, CSM); MBA• Was Senior Manager in Big 6 consulting• Head of Kitty Hawk Consulting, Inc. since 1991• Head of LeanAgileTraining.com• Started trying to do [Agile] before reading The Mythical Man-Month

– http://agileconsortium.blogspot.com

– [email protected]

2


Basics

3

© Joe Little 2013

Key Issue

4

Having better lives (the good life)

Smaller version: Getting more from Scrum.

© Joe Little 2013

Would it be good to have more Business Value?

5

© Joe Little 2013

What’s in it for you? (Scrum)

More business value for the firmMore for the customers

More, cheaper, faster, betterFaster delivery (TTM)Better way to work (for the workers)More funIt just makes sense

6

© Joe Little 2013

If you are a Team member...

If you deliver more real business value, your life has a higher purpose....

You have done something for another person(s). This is a wonderful thing.

7

© Joe Little 2013

What is the problem?

We are not delivering enough Business Value.

Large success does happen often...why not much more often??

And a larger success more often...

8

© Joe Little 2013

Put another way

We need more...InnovationInventivenessCreativityBeauty

WOW factorSex appealNeat, cool, wonderful solutions to hard technical and business problems

9

© Joe Little 2013

Per Tom DeMarco

Project A will eventually cost about $1 million and will produce value of about $1.1 million

Project B will eventually cost about $1 million and will produce value of about $50 million.

What do we learn from this?

See: “Software Engineering: An idea whose time has come and gone”, by Tom DeMarco

10


What we do now in Scrum

11

© Joe Little 2013

What does Scrum do ‘out of the box’ to get more business value?

1. Business and Technology collaborate2. We have a Product Owner for each Team3. We have a ‘prioritized’ Product Backlog4. Ordered mainly by Business Value5. Each sprint we have working product (per the

DOD)6. Each demo, we show the working product to

‘business stakeholders’ and get their feedbackAs in: ‘Did we build the best possible stuff for you this Sprint?’

7. We release earlier12

© Joe Little 2013

But we can do more...

13

© Joe Little 2013

What else can we do?

In small teams, with stickies. And large pens. One items per sticky...

What are some concrete things we could start doing tomorrow that could raise BV?

You have 2 minutes. The Team with the most stickies wins....

14


3 More Ideas

15

© Joe Little 2013

The 3 ideas

16

Business Value Engineering

Priority Poker

The Pareto Idea

© Joe Little 2013


A framework for continuously improving how you deliver more and more business value.

Steals from Value Stream Mapping (Lean)

Steals from Deming (Plan-Do-Check-Act)

Uses ideas similar to the scientific method

17

© Joe Little 2013

Priority Poker

Similar to Planning Poker, which you know and love

Puts BV Points on each fricken user story

Done as a Team

‘It’s the conversations, stupid.’

18

© Joe Little 2013

The Pareto Idea

We know this as the 80-20 rule...although not well enough.

Wilfredo Pareto: The ‘vital few’

Sifting to separate...The gold, platinum, diamonds

The silver and copperThe dirt (even the dirt gives a decent return)

19

© Joe Little 2013

Ummm...

These 3 ideas could be done separately....

OR...

They could be done together...

20



21

© Joe Little 2013


A framework for continuously improving how you deliver more and more business value.

Steals from Value Stream Mapping (Lean)

Steals from Deming (Plan-Do-Check-Act)

Uses ideas similar to the scientific method

22

© Joe Little 2013

The two essential questions

How do we get the BV ideas and the requirements into the Team?

How do we deliver the ‘product’ to the customer?

23

© Joe Little 2013

The BV process is visible and we articulate the underlying

Do you understand yours, end-to-end?

CustomersExternal

&

Internal

The BusinessCustomer facing

people

Internal groups (Firm oriented)

The Team

24

© Joe Little 2013

Hallmarks of good BV Engineering

1. The process is visible and articulated & always improving

2. Failures in BV communication are identified and corrected frequently, quickly

3. There are many theories, and a concerted attempt to prove out each theory

4. There is appropriate dynamism and change5. Business & Technology are partners6. Success is forecast (modeled) and also measured

after the fact7. Human judgment is involved (the numbers are not

a dictator)

25


Some good Theories (examples) - 1

The customer will change her mind 20% in 6 months. The customer does not really know what he wants.The customer cannot explain clearly what she wants.The customer only knows it when he sees it.The customer does not want software, just a solution to her problem.The Sales guys can explain some customer wants.

26


Good Theories (examples) - 2

We can learn something about BV using better metrics (maybe even money).While we need some documentation, it is also very important to give the Team the Tacit knowledge.The telephone game effect should be minimized. Let’s have real customers come to some demos.

27



Customers disagree. And customers and our shareholders disagree on what is highest BV. We are always separating the diamonds from the dirt. It helps for the Team to understand BV, even the monetary return from the new product. We expect to revise the NPV estimates over time, for many reasons.

28



It is helpful to identify different ‘roles’ for the product. We should optimize delivery ‘end-to-end’.“End-to-end” starts when the customer has an idea, and ends when the customer is successfully using the product (achieving their business value).

29



There are important benefit-cost trade-offs in our work. Only by having Business-Technology as partners, can we make the better trade-offs. Knowledge about these trade-offs evolves over time.Understanding knowledge creation and knowledge decay are key to achieving ever higher business value.

30



Every customer has a high demand for simplicity and speed of delivery. And high quality....even if unstated.Some projects will be cancelled if the benefit-cost ratio declines enough. Every representative of ‘the customers’ is problematic to some degree. We are always looking to reduce the bias.Let’s let the coder talk to some of the end users.Testers must understand business value more.

31



The business value of the (next) release can often change significantly in a month or two. (While we are building it.)Often customers need our help in prioritizing the “requirements” that the diverse people from that firm have given usThe best possible feedback comes after it goes into production. (release sooner)By using working software, we will learn faster how much we are misunderstanding what they really want.

32



Watching a person working if often the best way to identify the real needs.Use cases can be helpful in articulating the requirementsShowing some real users working software frequently can be very useful for learning about requirementsThe customer often does not understand his problem, and is more often far from articulating well the ‘better’ solution for it. (eg, does not understand what is possible)

33



Exogenous variables (war, weather, economic) add more reasons to deliver something smaller faster.No one person at the customer firm ‘knows’ what the customer (firm) wants. We need to re-train some customers to trust us enough to want more frequent releases. And give them the necessary support (eg, for testing).We must have higher quality so that customer testing costs and ‘pain’ reaches an acceptably lower level.

34


Some Outputs

A BV ‘process’ map (~30 objects/stickies)A list of underlying theories (~20)An ‘operational’ definition of BV for the next releaseA BV ModelA specific improvement to make (later: approved & work done) Something changesA way(s) of measuring (higher) BV (eg, after the release is in production)A way of evaluating success (later: success evaluated)

35


Priority Poker

36


What is it?

Priority Poker is very similar to Planning Poker, except it is about Business Value.

Other differences:The reference story is the highest BV one. (Not the lowest effort one.)We use the “5” best “experts” on Business Value.

37

3 roles• Product owner• Scrum master• Team

3 artifacts• Product backlog• Sprint backlog• Sprint burndown

4 activities• Sprint planning• Daily scrum• Sprint review• Retrospective

CSM v9.3 © Jeff Sutherland 1993-2008; © Joe Little 2013

Planning poker cards

http://planningpoker.crisp.se

Source: Henrik Kniberg38


What happens?

They select the reference story. BVP=100They choose a story (random) to evaluateThey discussThey vote. Imagine: 2, 20, 20, 20, 40The two extremes talkThe vote again: 13, 20, 20, 20, 40. The average is 23. (Meaning: On average, they feel it is about 23% of the BV of the reference story)They go to the next story

39


What is the pupose?

The number of each story card?

OR....

That the 5 ‘experts’ discussed some important issues, and ‘everyone’ now understands BV much better?

40


The purpose is...

...that now everyone has shared their (key) tacit knowledge about business value.

...that now everyone has learned.

...that now mostly they see the same elephant (of BV).

41


Now...

R = BVP / SP

We can do more work on the Pareto Idea... (up soon)

42


Very simple...

...very powerful.

43


A word of advice...

If the ‘wrong’ person explains it, they won’t do it

If the ‘right’ person explains it, they do it easily and see the value

Hard to explain why...but that seems to be the truth

44


The Pareto Idea

45


The Pareto idea

80-20

The vital few

Less is more...

Example: The iPod, not the Zune

46

© Joe Little 2013

Pareto’s Rule (you must prioritize!)

47

0

20

40

60

80

100

1st Quint. 2nd Quint. 3rd Quint. 4th Qunit. 5th Quint.

Pareto Curve

© Joe Little 2013

Pareto said...

You can re-apply the Pareto idea at every level of the population

THUS:

Every time you break down an Epic, part of it is high value, part medium, part low

48

© Joe Little 2013

Advice

Smaller stories!!!!!!

OK?...

Smaller stories!!!

49

© Joe Little 2013

More importantly...

You have been brainwashed, for years, into the 100%-100% rule

And everyone around you has been brainwashed

AND... we are still learning how to sift & see:gold, platinum, diamonds

silver, copperdirt

50

© Joe Little 2013

Advice: Learn to say ‘NO’

...in a nice way

See: The Power of a Positive No, by William Ury

51

© Joe Little 2013

Advice: First do the 85%-50% rule

85% of $3 million is $2,550,000

50% of 12 months is 6 months....

THUS: a 70% improvement with just the 85-50 rule

52

© Joe Little 2013

Advice: Work on it every day

Work on it HARD every day

Every day there is some additional work (feature) you can identify NOT to do in the Release.

Identifying that is hard.

Saying NO is hard.

53


Killing Babies or Sizzling Steak

54

© Joe Little 2013

You have finished Release Planning...

55

By Apr 30th!

P.B.

© Joe Little 2013


56

By Apr 30th!

P.B.

NO !!!

© Joe Little 2013

“Killing Babies”

I don’t recommend it

57

© Joe Little 2013

Show a Pareto Chart

58

0

20

40

60

80

12

34

5

R Factor

Bar Chart - Stories - Width is SPs

© Joe Little 2013


59

By Apr 30th!

P.B.

© Joe Little 2013

Sizzling Steak!

Yes! We have to show the sizzling steak!

60


Other basic ideas

61

© Joe Little 2013

Some simple things - 1

62

Benefit-cost analysisLet the coder talk to the real userAgile SpecsAsk the Doers to think about business value frequentlyCreate a BV ModelModify the BV Model once per monthLearn to cancel projects when better projects come along

© Joe Little 2013

Some simple things - 2

Identify & fix more technical debtMake the PO betterGet more PO time for the teamTeach the PO about the productGive the PO time to talk to end users and customersLet the PO watch a master PO Improve the quality of the Agile Specs

63

© Joe Little 2013

Some more simple things - 3

Get more creativeMake the product more elegant or beautifulGet better business stakeholdersMore feedback from the business stakeholdersMore time from the business stakeholdersA better relationship between the business stakeholders and the POBring real end users to the Demo

64

© Joe Little 2013

More simple things - 4

Map the stories (in several possible ways)The Doers should watch what the end users do each dayRe-calculate the BV Model more frequentlyIntegrate more between Team planning and firm planningStudy competitive productsRemove small features from the release (Pareto)Faster feedback within the Sprint (Doers - PO)

65

© Joe Little 2013

More simple things - 5

Have more fun!?!***Ask: “What will delight the customer?”Measure BV delivered (measure after a release)Focus on TTM (time to market). Measure it.Identify the BV drivers specific to your effortClarify the acceptance criteria (functional tests) moreAutomate the functional tests (more)

66


Summary

67

© Joe Little 2013

Do these!!!

68

Business value engineering

Priority Poker

Execute on the Pareto Idea

© Joe Little 2013

If you do, you’ll have more fun!

69

Yes, even hard work can be fun, if you do it the right way....

© Joe Little 2013

Thank you.

Please contact me here:

Joseph Littlejhlittle@kittyhawkconsulting.comLeanAgileTraining.comLeanAgileTraining.com/blogTwitter: jhlittleOffice: 704-376-8881Cell: 917-887-1669

70