Seminar: Dial “M” for Metrics: Agile Metrics and Data Science Full Day

Posted by in Announcements, Events |

We are proud to work with great partners. In this power-day of Agile metrics and analytics, Focused Objective’s Troy Magennis has teamed up with Larry Maccherone, formally of Rally Software where he led the team responsible for their metrics and analytics features.

This first seminar will be held in Seattle, 14th October 2014. This seminar is very competitivly priced, and you would rarely get two industry leaders for one day at this price. Please book early because we are limiting the number of attendees to ensure quality interactions with attendees.
Eventbrite - Dial "M" for Metrics: Agile Metrics and Data Science Full Day

 

Dial “M” for Metrics is a one day seminar that covers the entire lifecycle of an Agile metrics process, from metric selection through to reporting data in compelling ways. After this seminar, participants will be well versed in all aspects of Agile metrics and basic data science for use in their own IT organizations.

The day is structured with multiple lecture style presentations from industry experts on all metric topics from selection, capture, analysis, forecasting to reporting. The presenters are all experienced in the practical application of metrics for IT organizations from team to portfolio and organization perspectives.

This seminar is suited to you if you –

  • Struggle to know what metrics are useful and which ones are misleading
  • Suffer from data quality through poor capture and gamed values
  • Find analyzing and presenting data to get action is harder than it ought to be 

Target Audience –

  • Executives or mangers wanting a better understanding of Agile metrics and analysis
  • People responsible for Agile project planning and reporting
  • People interesting in expanding their knowledge on Agile metrics and analytics who are using Scrum, Lean, Kanban or ScrumBan as their IT process

Topic areas –

  • Metric Selection – picking the right metrics that add value and avoid gaming or poor performance
  • Metric Capture – identifying errors and cleaning noisy or gamed data
  • Metric Analysis – forecasting and interpreting variation in data that is significant
  • Metric Presentation – presenting data to get action and avoiding common presentation mistakes
  • Making Analytic Decisions / Q & A – wrap-up combining all the practices into a useful management program
 
Coffee povided (it is Seattle). Lunch will be on your own at one of many downtown locations. We suggest joining your colleagues for networking time or joining the speakers for vigorous discussions. This is not a sales conference. This is a training course with lots of time for specific questions. The speakers will be available during the breaks and specifically at lunchtime for one on one advice. 
 
Detailed Agenda
Subject to change on the day based on audience feedback!

Time

Topic

Details

90 min

Metric Selection
by Larry Maccherone

– Seven Deadly Sins of Metrics

– Metric Selection Strategies: ODIM/MIDO

– Big Data Science versus Tactical Metrics

– Value of Different Metrics vs Cost of Acquisition

60 min

Metric Capture

by Troy Magennis

– Capturing Data Correctly

– Cleaning Data from Errors and Bias

– How Much Data is Needed

– Capturing Data Context

90 min

Metric Analysis

by Troy Magennis

– Analyzing Data Integrity (error checking)

– Analyzing Variability

– Forecasting and Prediction

 – Regression

 – Monte Carlo

– Limits of Certainty

– Sensitivity Analysis

60 min

Metric Presentation

by Larry Maccherone

– Presenting accurate and compelling data

– Lying with Statistics and knowing others have

– Comparing Data Across Teams Safely
– Presenting uncertainty in results

90 min

Making Analytic Decisions / Q & A

by Larry and Troy

– Starting Organization Wide Metric Programs

– Quantifying the Impact and ROI of Metric Programs

– Industry Progress – What tools do things well

– Where does Big Data fit with Agile

 

Forecasting Hands-on  Tutorial (optional, runs Wednesday)
Agile Forecasting by Troy Magennis. This will be a hands-on day building probabilistic forecasting models for Agile (Scrum, Kanban, ScrumBan) teams and projects. PC or Mac running Windows in emulator if you want to follow-along during the exercises. By the end of this day attendees will be well versed in building probabilistic models and forecasting their own IT projects.
About the Speakers
Larry Maccherone

Larry is an industry recognized Agile thought leader. Larry most recently served as Rally Software’s Director of Analytics and Research where he led a team using big data techniques to draw interesting insights, provide software development performance metrics, and provide products that allow Rally customers to make better decisions with data.

Before coming to Rally Software, Larry worked at Carnegie Mellon with the Software Engineering Institute for seven years conducting research on software engineering metrics with a particular focus on reintroducing quantitative insight back into the agile world.

Larry is an accomplished author and speaker, presenting at major conferences for the lean and agile markets over the last several years, including the most highly rated talk at Agile 2013.

Agile 2014 Interview

Troy Magennis – Focused Objective LLC

Troy has been involved with technology companies since 1994, fulfilling roles from QA through to CTO over the last 20 years. Troy speaks at many Agile conferences and has played an Agile training and mentoring role for executives in small and large organizations (Walmart, Microsoft, Sabre, Siemens being previous clients).

Troy currently consults and trains organizations wanting to improve decision making on software portfolio and project questions through Agile and Lean thinking and tools. Applying Scrum and Lean techniques appropriately and where they are going to make this biggest benefit through quantitative rigor.

Agile 2014 Interview

Agile 2014 slides

Interview “Projects at Work”

 

Read More

Agile 2014 – Software Moneyball

Posted by in Events, Featured, Forecasting |

We were honored to be selected to speak at Agile 2014 on matters close to our hearts – Metrics, analytics and forecasting. This session focused on the reasons why selecting the right metrics is important and also how to quickly assemble the right metrics into reliable forecasts.

Agile 2014 Interview

Agile 2014 slides

Interview “Projects at Work”

“Metrics help forecast and manage Agile projects and teams. Used properly, simple statistical techniques add insight and encourage better actions to be taken. Velocity is the general go to metric, but how well it forecasts the future is the elephant in the Agile community’s room. This session talks about how to capture metrics (other than velocity), how to identify which metrics cause the most impact when changed (carry the most information), and how to use metrics to probabilistically (statistically) forecast a future. We briefly discuss #NoEstimates, and this session touches on the need to move beyond story point estimation in order to improve analytically.
This session will give attendees practical techniques to apply on their own teams and projects, and confidence to talk about forecasts and uncertainty to others. It debunks the common myths that you need lots of data, a Phd in Mathematics and that you need historical data to begin with.

Topics explored include

* Understanding Uncertainty: example problems to help understand how poorly humans assess uncertainty
* Common Errors Traps: why you need to forget standard deviation, and common data mistakes in our industry
* Capturing Data: how to capture data cleanly and prepare it for analysis
* Statistical Forecasting: how we use the data we capture to predict and assess
* Sensitivity Testing: finding what matters most, and why its not story point estimates!
* Waterfall, Scrum and Lean/Kanban Metric Profiles: history of project processes and how they impact forecast-ability and uncertainty (the dreaded cone of uncertainty debunked)

Showing data from over forty teams, this session will show how common patterns (Scrum versus Lean, Operations versus Development, Waterfall versus Scrum) are evident based on team structure and style, and how understanding these assumptions can make managing teams more productive.

Why Software Moneyball? I call a quantitative approach to software project management, “Software Moneyball.” The book, “Moneyball” describes how the Oakland A’s baseball team used analytics to beat better funded competitors ($40M versus $120M), and transformed the business of baseball. If you compete against better funded rivals, you must outsmart them to succeed and better use of analytics is often the sharpest tool in the toolbox. If baseball Managers can do it, so can you.”

Read More

Cycle Time Mathematics – KLRAT Presentation

Posted by in Events, Featured, Forecasting |

I had a great time at the Kanban Leadership Retreat held in Mayrhofen in Austria. This is a un-conference held with Kanban practitioners to present new ideas in Kanban to leaders of that community.

I presented some new thoughts around cycle time mathematics. How and why reducing the variation in cycle time impacts forecast certainty. I also expanded this into some techniques to identify what areas of delays and blocking events give the biggest reward in reducing cycle time variation by analyzing both the frequency and the duration aspects of queuing and blocking event. I’ll be more fully writing this up soon.

Here is a link to the presentation: Cycle Time Mathematics – KLRAT Presentation

Note to attendees: This presentation contains a lot more reference material than I covered in the 45 minutes!

Troy.

Read More

LKNA 2013 Conference Tutorial – Lean Forecasting: Speaking Value Rather than Cost

Posted by in Events, Forecasting |

We are proud to be attending and speaking at the Lean Kanban North America conference later this month in Chicago. As part of that conference we are running a full day tutorial on Forecasting.

 Here are our goals for the one day tutorial –
  1. How to forecast likely delivery dates and costs
  2. How to calculate Cost of Delay and why it matters
  3. How to quantify the benefits of decreasing cycle time in cash-flow and throughput
  4. How to calculate the cost and date impact of defects and blocking events
  5. How to analyze systems for staff imbalance
  6. How to analyze and use historical data when available
  7. What constitutes “enough” historical data and what to do when you have less (or none)

We surveyed the participants and the order listed above is the order the attendees picked. There is still time to register, and we have prepared our tutorial material to maximize one day of learning, with plenty of exercises to take home and continue learning.

Here is the description –
 “Humans stink at planning through uncertainty, but that doesn’t stop executives, their managers, and financial departments from expecting and acting on estimates fraught with more uncertainty than confidence. Despite this, most project teams go through the ceremonial and demoralizing “estimation” process knowing their estimates are worthless at best, and dangerous at worst. In fact, estimates are so untrusted by everyone that executives, managers, and financial hawks routinely dismiss them, make arbitrary changes and hold the estimates (and those who create them) with utmost disdain.

What can be done about this?

How about forecasting instead of estimating?

What’s the difference? Estimating is typically made up of little more than a wet-finger held to the air. (A bit comical, but not far off the mark.) Some wet-fingers are well-calibrated, but the calibration doesn’t change some of the fundamental flaws of estimation. Foremost of these flaws are the flaws pertaining to the rationale behind the estimates and the ratio of knowns to unknowns to unknowable that common estimation techniques ineffectively handle.

On the other hand, forecasting is based on actual system capabilities. As importantly, forecasting can be tooled to speak to executives in the terms they already use throughout the business — except when it comes to software or IT services. Terms that speak in the language of value rather than the language of cost.

In this tutorial, attendees will learn surprisingly uncomplicated approaches to understanding system capabilities and use this understanding to express the need for resources, time, and process improvements that are quantitatively solid and statistically valid. In other words, attendees will learn how to make a case for staffing, resources, time, and even contractual commitments using basic tools, a little data and a helping of visuals in the forecasting language of executives rather than the estimation language of charity cases.”

hope to see you there.

Read More

Training Course: Effective Modeling and Simulation

Posted by in Announcements, Events |

We are pleased to announce a new training course to accelerate understanding and to reduce the time to being effective.

Effective Modeling and Simulation (2 Days)
This 2-day training course gives the participants an intermediate to advanced grounding in modeling and forecasting software development projects. This course is based on our simulation products and the Forecasting and Simulating Software Development Projects: Effective Modeling of Kanban & Scrum Projects using Monte-carlo Simulation book, this hands-on course gives more background and advanced lessons that mean forecasts will be more reliable and accurate. Suitable for Scrum and Kanban teams (tailored for the audience).