Monday, February 11, 2019

Boss, get out of the way

“Could you imagine if school teachers allowed students to pick their own grades?”

This was Boss’ response to the idea that agile teams can plot their own “maturity” against an agility rubric. Boss told me that the agile coaches should go around with a clipboard, citing what each team is doing right and not right against the rubric.

My response to his response was basically, “you can go ahead and use this agile maturity exercise to rate and compare teams, but don’t ask me to do it. And, while you’re at it, I don’t suggest you do it either.”


Many years ago, I was a school teacher (high school and junior college), and I did what Boss said should never be done. At grade time, I would take each student aside and ask the two simple questions: “What grade do you think you deserve?” “Why?” Compounding the initial confusion was the condition that I hadn’t assigned any number or letter grades during the term. Everyone’s first assignment received a check (√), no matter what. I told them when handing back the assignments that this was a baseline, and that the name of the game was improvement.

After the first assignment, everyone was judged against themselves, not their peers. So,
Progressing as expected à
Outstanding progress à √+
Blew me away à
Not impressive à √-
Didn’t do the assignment, or did a half-assed job à 0

The nuns at the Catholic high school didn’t like my system much more than Boss did. I wish in retrospect that I had the perspective now to explain it to them.

Here are some of the findings (n~=200 students over 2 years):
  • The vast majority (80%) of students suggested a grade that I could live with, that roughly represented their progress and effort. These were easy. “OK, 85 sounds about right to me. Let’s go with that. If you continue to get better throughout the year, that grade will go up. If not, it will go down. How does that sound?”
  • ·        Almost no students asked for a grade that was wildly too high. In response to the 1-2 times that a student who put in a lackluster effort asked for a 90+ grade, a simple “really?” from me got the conversation back on track.
  • ·        A good chunk of students suggested a grade that was WAY too low.

In those cases, when I asked (what I thought was) a 90 student why they thought they deserved a 75 for a grade, I got an instant teachable moment. Most of those kids suffered from low self-esteem or a self-fulfilling prophecy complex.

Here’s a typical conversation:
“I’m bad at English. I always get around a 75. I’ve never gotten an 80.” one kid told me.
“Why do you think that is?”
“I’m not such a great reader or writer. And usually the material is really boring….”
“That’s very surprising to me. I’ve seen you work really hard this term, turning in all of your assignments on time. And, you’ve gotten a lot better as a writer..”

And so it goes.


This agile maturity exercise is meant to be a development activity, intended to help individuals or teams to get better. It necessarily needs to be separate from performance evaluation activities. If you, Boss, conflate development and performance, bad things happen, and you,
  1. Undermine empowerment. By telling the teams what they are doing right and wrong, you are telling them that they are unable to do same.
  2. Sow distrust. If the teams think they are being rated and compared, they will, in the interest of self-preservation, do things to undermine the system.
  3. Show that you don’t think much of their ability to make decisions. Let’s face it: Your people can read a rubric and map their own performance against it. If you can’t trust them to do that honestly, then that’s a big problem.

If we coaches aren’t going to rate the teams, what should we do? The overall role of the coach is to help teams perform better. We do this in different ways. Sometimes, we share our vast agile knowledge (teach). Other times, we lead individuals or teams through a process (facilitate). Occasionally we directly advise individuals and teams on how to get to the next level of performance (mentor). And often we ask questions and lead people to find answers to their own problems and dilemmas (coach).

The art of coaching is to figure out when to assume which stance at which time, keeping in mind that the performance and the outcomes belong to the team. The coach is not part of the team.

Now, no matter how empowered (or whatever) the team is, the Boss still employs them. But, we’re not talking about hiring and firing here; we’re talking about performance. And, to get the best performance out of a team the Boss needs to tell them what good performance looks like, and then get out of the way.

Friday, January 4, 2019

The Answer to Your Process Question? It Depends!

In my role as an agile coach, presenter and facilitator, people ask me process questions about agile methods every day. After all, I am an agile “expert.” This is what people expect me to know.

“Should we assign story points to a spike?” goes one such typical question.

“It depends…” begins the answer that I always want to give. Of course, if I start out every response with “it depends,” it will get annoying. What I mean is, the answer to the question depends on who you ask, and more importantly, what you’re trying to accomplish with this decision. I want to facilitate the questioner through the thinking, so they can answer their own question, armed with the conventional wisdom.

Here is a simple technique to try.

Step One – Answer the Question
Respect the question and present the generally-accepted consensus answer, if there is one. If there is no consensus, I cite both sides of the argument. “Well, that’s an interesting question. One the one hand, reason ABC says you should assign points. On the other hand, reason XYZ says you shouldn’t….” Teasing both sides cues up a meaningful discussion.

Step Two – Probe
I’m trying to set up a teachable moment here, and I use questioning to get there. The lesson is that you should follow or not follow certain rules based on how that rule will serve your team. There are a number of questions that can lead you there - What do you do now? What doesn’t work about how you do it now? What is the problem? Why are you even talking about this?

I like to try a few of these questions to get to the underlying issue. Then explore the issue, trying to facilitate the person’s (or team’s) understanding. Whatever decision is made should help the team to be better, and whatever happens should be retrospected along the way. E.g., you might start assigning story points to a spike, and it doesn’t fix the problem, or it causes some unintended consequence that needs to be addressed.

Step Three – Revisit the Question
My questioning leads the asker of the question to a better framing of the issue. Now, I can be the asker. “So, based on all of that, what do you think you should do?” There are countless directions this could go, but this type of examination is better than just taking the answer out of a book or out of the mouth of a so-called expert.


While many cast me as an agile “expert,” my ultimate value is in helping people (individuals and groups) discover best answers to their own questions. Ultimately, I want to move them to a process where they examine the impetus for the question.


Q: Should you assign story points to a Spike?

A: It depends. If it helps you to work better, then you should.

Tuesday, September 25, 2018

Mr. Medoff

I - The Lesson

Mr. Medoff sits comfortably, legs crossed in his grey suit, at a paper-strewn desk, smoking a cigarette. A few feet away, I sit at his grand piano. It is the last few moments of the last piano lesson before Christmas. I ask a question about theory.

“How do you know what key a piece is in?”

“What do you think?”

“Well, is it the first chord of a song?”


“Is it the flats and sharps in the sheet music?”


“Does it have to do with the chord sequence?”


“Um. I don’t know then.”

“You think about that, and we’ll talk about it next time.”

With that, he adds up the letter grades he gave for each of my lesson performances, and congratulates me for an A- overall for the day. He dodders across the room and sticks his head into a small storage closet.

“Do you have Mozart yet?”




“How about Schubert?”


He emerges from the closet and hands me a miniature vinyl bust of the Austrian, my eighth such prize in 18 months of study with him. I’m happy because it will create lovely symmetry at home, four on each side of my piano sill.

II - All Lessons Are Cancelled

Two weeks later, my mother hands me two quarters for two subway  fares and a $6 check for that day’s lesson. I board the T at Government Center and head out to Mr. Medoff’s Brookline home.

My mother was steadfast in her desire for me to play the piano. She wanted to take lessons as a child, but couldn’t because they were poor. So in first grade, my folks bought an Acrosonic spinet piano, and I started lessons.

As I make a snowy five-minute walk up the hill from the T stop, I ponder the question from the previous lesson. I have no new answers. If sharps and flats don’t determine what key a song is in, I am eager to hear what Mr. Medoff has to say on the matter.

To enter Mr. Medoff’s studio, you have to walk down a side walkway and use the lower level entrance in the rear of the house. I stop and read a handwritten message that greets me from the storm door:

“Due to an unfortunate circumstance, all lessons are cancelled.”

After a moment’s consideration, I turn my ass around and take the train back home.

III - Resolution

My mother’s reaction to my story is measured. “There must have been a funeral or something. Just go back there next week.”

The next week I repeat the drill, with two new quarters and the same $6 check. On the storm door the same note greets me:

“Due to an unfortunate circumstance, all lessons are cancelled.”

Mom calls Mr. Medoff’s house. His wife answers. “Oh, Arthur died.”

Taking a seat in the comfy chair in the den, I mourn.

IV - Arthur Medoff

A native of West Roxbury, MA, Arthur Medoff died on December 26, 1975 at 55 years of age. I could have sworn he was older. But hey, I was 11.

Four decades later, Arthur’s memory lives on the internet. One of his former students, now a piano teacher, cites his influence on her professional profile. You can find Arthur’s magazine writing, a column called Interlude for the Boston Musician’s Union and book reviews for American Music Teacher. He had four children with Evelyn, who outlived Arthur by 31 years.

V - The Question Remains

Through the years, I’ve asked musician friends to answer the question, “how do you know what key a piece is in?” Nobody has given me a satisfactory answer, because every idea staggers after we discuss paradoxes or debilitating edge cases. The three responses I gave to Arthur make some sense, but fall apart under the slightest scrutiny.

The internet provides no consensus, just a bunch of overly complex stabs that ultimately miss the mark.

I consulted my copy of Arthur’s 1956 book called Fundamentals of Style in Popular Music (not available anywhere). Rich in theory, it never directly addresses the question.

Musicians ask no questions if told a song is in the key of F, yet nobody has a pithy definition for this process. 

Everyone knows, but nobody can explain. Maybe that was Mr. Medoff’s point?


Image references

References for the biographical information

Friday, September 14, 2018

Providing 360 Feedback Without Anyone Getting Hurt

“We're going to turn this team around 360 degrees.” (1)
-Jason Kidd, basketball player and coach


It happens from time-to-time, during performance evaluation time, or when a new “360 feedback” event kicks off. You start receiving any number of requests for feedback about your peers, subordinates, or bosses.

360-degree feedback (aka multi-rater feedback) is a potentially powerful tool, gathering balanced information about employees for development planning or performance appraisals. While the intentions are noble, it shares many of the characteristics of other heavyweight, centralized, time-wasting, stress-inducing, HR-driven processes. If misused, and it often is, 360 feedback adds up to one gigantic antipattern.

For most people, you will have more immediate success working within the system than you will complaining or trying to change it. Here are some suggestions:

Apply radical transparency

The process calls for you to fill out a form or answer questions and deliver your feedback (now called ‘data’) it into some anonymous machine, which delivers the feedback to the employee anonymously or via the manager. It’s scary, especially if some of your colleagues are inclined to weaponize the anonymity of such a process.

Before you feed the system, apply radical transparency. Have a discussion with that person, and simply put your notes from that conversation into the tool. Share the written submission with the person, by copy/paste from the tool or copying your email response.  Hopefully this radical transparency will improve the working relationship you have with the subject of the feedback.

Subvert the performance appraisal aspect

Sometimes companies or functions use 360 feedback to differentiate performance, feeding data to the annual forced distribution, norm-referenced performance management machine that is so demotivating. Don’t conflate performance with feedback. Approach the activity with the mindset that you would any other feedback, being helpful, encouraging and non-judgmental. If you are forced to give numbers, use grade inflation: Only 4s and 5s (on a 5-point scale) to give some differentiation within the categories without any lower marks. If you give feedback to multiple people, make the numbers add up the same for each person.

Create a teachable moment

If you’re in an environment that is grumbling about the feedback process, explore it openly and figure out ways to make it better. Ask: What is it about this process that is so infuriating, stressful, unfair, time-consuming, or <whatever other emotion>? What can we do to alleviate that? What do we hope to get out of such a process? What can WE do to make it work better for us?


For those organizations ready to change, what you land on is this: Eliminate the event-driven feedback process, and create an environment where feedback is ubiquitous, helpful, and welcomed. Make your workplace one where everyone can help everyone to get better, and everyone can win.

(1) Quote is possibly apocryphal, but too delicious not to include.  See: 

Thursday, August 2, 2018

MAILBAG: How to Define Success for Your Agile Initiative

From the mailbag:

Dear @TheBobWinter,
Our IT department is shifting from the traditional Waterfall approach to Agile. This will not only impact IT employees, but anyone within the company who works with IT on a project. We have a training plan, and employees will be educated via workshops on how to work in an Agile environment. Now, our  IT leadership wants to know:
·        How will we judge the effectiveness of the training?
·        How will the shift to Agile impact the business?
I would look at for the answer in your book, but I am on the road this week.
Scott in Scottsdale

Scott in Scottsdale,

Nice to hear from you! The answer to your questions could and does fill countless books and articles – including Chapter 6 of my bestseller Agile Performance Improvement.

It sounds like you are cooking up a measurement plan to accompany your learning strategy. Here are some starter thoughts. Proving the efficacy of an agile program is tricky, because you can’t necessarily make direct connections between the training or the PMO efforts and the business results on the other end. However, you can demonstrate some good value in these three categories.

If you look up “agile metrics,” there are all sorts of ideas in the general category of ‘conformity to practices’. So, are the teams doing x practice (stand ups, planning, retrospectives, etc.)?  At some point management will probably want to know if teams are actually “practicing agile,” so they need to decide what that threshold is in terms of, say, “%age of adherence to practice.” Once your teams reach higher levels of maturity, these numbers will become less meaningful, since teams will learn that not all practices work for everyone.

Business people know that these agile practice indicators don’t pay the rent, but they typically do want to be able to say, “our people are doing agile.”

Team Performance
More importantly, the business will care about whether key indicators of team performance have improved in a measurable way. Typical metrics would be things related to speed (e.g., velocity or delivery rate) or quality (defect rate) or customer satisfaction (NPS).
The business should articulate what they are hoping to improve before kicking off the agile initiative, and a critical few metrics should be tracked. The underlying assumption is that whatever the focus on team performance, it has strategic alignment with what the business is trying to do overall.

Employee Engagement
Some executives consider it “fluffy,” but employee sentiment can be a powerful indicator of how things are going. Set a baseline and monitor employee engagement, both in general and with the Agile initiative. Research shows that engagement (or satisfaction) is positively correlated with discretionary effort and individual performance.


Within each of these categories, there are countless possible measures. I recommend choosing a small number that will resonate with the key stakeholders. As such, any type of plan of this nature should be codified only after consultation with the business people. After all, they are the key stakeholders, and they should be on the same page with the desired outcomes.

The whole idea is: Are we getting better? How will we know that we’re getting better? Do the indicators represent what is important to the business?

I hope this helps. Let me know how it goes, and keep in touch.


Friday, April 6, 2018

Planning Your Own Professional Funeral

NOTE: This entry originally published on the Enterprise Agility blog at CA Technologies. 

As an agilist, the further I progress, the more I feel like I’m planning my own professional funeral. That is, if enough good things happen, I could be out of a job. If teams become truly collaborative, the less they will need agile coaches. As teams become self-directed, they won’t need so much formal training. The same is true for others in the corporate workplace: As teams become empowered, they won’t need managers to manage resource allocations. As teams of teams become better aligned, they won’t need project managers organizing everything around them.

For a lot of us, the better job we do, the less we are needed.

Well, let me correct that: The better job we do, the less we are needed to do the things we currently do. So, then what?

Looking back, there are many jobs out there that have either outlived their usefulness or dramatically changed due to technological evolution. ATMs have changed the role of the bank teller. Planning and communication tools have changed the role of administrative assistants. POS and security systems have changed the role of gas station attendants.

These changes don’t necessarily mean those people lose their professions. For example, the bank teller profession is alive and well.

However, as rote duties become automated, workers need to be redeployed to do more valuable work. This, of course, requires skill development. Tellers learn sales skills; admins refine their facilitation skills; gas station attendants manage food stores.

In creative enterprises, the same will be said of the office jobs listed at the top. As companies succeed in creating “learning organizations,” preparation for more value-added, un-automatable contributions will just happen, only if…

If leaders commit to creating an environment where learning and empowerment is encouraged, everyone can succeed together. Leaders, do you believe that people are resourceful and will seek the learning they need to succeed? Do you believe that within collaborative teams, peer teaching can happen just in time? Are you ready to pry your gaze away from required training content, mandated “best practices” and competency-based development? If the answers are yes, you might be on your way.

For those who would plan their own funeral, it’s time to plan for your professional rebirth. Ask yourself: How much value am I adding in what I do now?  What do I need to do and learn to make my team a success? Is it possible for me to thrive in the environment where I work now?

Listen to the answers to these questions. Listen to yourself.

Thursday, March 8, 2018

Why Gilbert Matters

Four decades after the original publication of Human Competence: Engineering Worthy Performance, let's consider the ongoing relevance of Thomas F. Gilbert’s most famous work. Certainly he remains the most revered and most important figure in the history of human performance improvement field. Does he still matter? 


When I was in junior high school, I was a paperboy. I usually chose to walk rather than bike my route so that I could read The Boston Globe for an hour. Here are some of the big news stories that stick in my mind from 1978: A great blizzard dumps 21” of snow on an unprepared Boston; Bucky Dent’s dramatic home run propels the Yankees to a division title over the Red Sox; Pope John Paul I dies after only 33 days of papacy. Whether or not you were following news back then, you can see that 1978 was some time ago.
The 40th anniversary of the publication of Human Competence: Engineering Worthy Performance is an invitation for reflection. ISPI folk celebrate Thomas F. Gilbert as the most revered figure in the history of our profession. After all he is ‘The Father of Human Performance Technology (HPT)’ and the namesake of ISPI’s highest honor. His Behavior Engineering Model (BEM) remains (possibly) the most cited model or concept in any gathering of human performance technologists.
Human Competence was first published before I entered the workforce, and Gilbert passed away before I had heard of him. After attending my first ISPI conference in 2008, I purchased a tattered copy of the Tribute Edition, which was published in 1996 shortly after Gilbert’s death. In this version, you get not only the book itself, but 17 pages of heartfelt messages that demonstrate what he meant to so many colleagues and friends. Gilbert’s unfinished posthumous autobiography, cleverly called Human Incompetence: Confessions of a Psychologist, adds 90 more pages of praise in its Addendum. The picture is clear: In addition to being a brilliant thinker, storyteller, and mentor, apparently, he was a great guy as well. I’m sorry to have missed him.
There is little question that Gilbert was great. Does he matter in 2018?
For answers, I returned to the text of Human Competence, with a reading that considers his relevance to the consultant or HPT practitioner of today. I conclude that yes, Gilbert still matters, and Human Competence is worth reading or re-reading today. Here’s why.

The Book Itself Is Great

“Length in perception has a certain independence of length as measured by a ruler. As we move away from a package of cigarettes, the length of its physical image on our retina grows smaller, nevertheless, the package seems 80 millimeters long at any distance.”
-Human Competence (Tribute Edition), page 339

Human Competence hits a rare sweet spot for a business book – somewhere between bathroom book and ponderous tome. It’s light (entertaining) enough so that it’s hard to put down, but heavy enough that you don’t want to read too much of it in any one sitting.
The quote above encompasses so much of what is distinctive, pragmatic, and charming about this book. Certainly, the reader will notice that Gilbert was a teacher at heart and professorial in tone. I find myself engrossed in his examples, immediately wanting to reread paragraphs or sections as soon as I finish them. This is the type of book that you might carry around in your bag for months, pulling it out during downtimes when you know it’s time for a break from your smartphone feeds. While there is a brilliant arc to the structure of the entire work, I find myself opening it up randomly rather than picking up where I left off. Sometimes it’s just enough to read some of Gilbert’s unique homespun verse that introduces most chapters.
The image of the cigarette package is a typical of the anachronistic turns you see from time to time. You’ll also see outdated business terminology. Hey, it was the 1970s. It’s 99% business in here, but Gilbert will take short asides on race relations, gender roles, or other contemporaneous issues. It gives the whole work a bit of a time capsule feel, which I find charming and amusing. Moreso than other books of its type and vintage, the author’s personality animates the work, and that is to the reader’s benefit.
As with this example, Gilbert’s lectures seem oblique when suddenly he makes a turn towards the subject matter at hand. Here Gilbert prefaces a discussion of the dimensions of human behavior with a discussion of the dimensions of the physical world. Side trips into scientific research methods and the relevance of animal behavior to the study of humans often underpin his messaging.

He Taught Us Timeless and Enduring HPT Lessons

“In the great cult of behavior, the appeal is to control or affect behavior in some way. There is little or no technology of ends and purposes. Indeed, behavior itself is view as an end rather than as a means to an end.”
-Human Competence (Tribute Edition), page 7
It's easy to say that Gilbert is outmoded because he's his best work is 40 years ago, and a lot has changed in the world. You will still find timeless nuggets of wisdom, some of which continue to be overlooked.

Lesson 1: Behavior ≠ Performance

Before we get through the first 10 pages of Human Competence, Gilbert introduces “the great cult of behavior,” and the image endures. He observes that managers equate performance with behavior at their own peril, and we all see the same today. This thinking leads to the still ubiquitous management anti-pattern: The assumption that performance is mostly a product of an individual’s knowledge, hard work, and intrinsic motivation. Who here hasn’t seen the manager who blames poor outcomes on incompetent or unmotivated employees instead of factors in the environment?

Lesson 2: It’s about the Environment

The most commonly cited lesson comes from the application of his Behavior Engineering Model (BEM): Factors in the environment have a greater impact on human performance than factors within the individual. Seldom does a week pass in my work where I do not need to correct a manager’s notion that remediating individual capabilities through training alone is going to have an impact on the problems their business face.

Lesson 3: Start by setting expectations

Gilbert sequences the six boxes of his BEM according to the likelihood of causing performance deficiencies. The failure to set proper expectations with employees is the number one contributing factor to poor performance. If people only knew what was expected of them and what outcome is expected of them, they would have a much better chance of doing a good job and helping to achievement of organizational goals.

He is Still Relevant for the 21st Century

 “To love is to stand in the human vantage point, which requires us to stand with others for a common purpose.”
-Human Competence (Tribute Edition), page 348
The idea of starting with a common purpose is as old a humankind. Ancient hunter and gathering people would not cite their purpose as having sharp tools. Their purpose was to keep the family from starving to death. Now in recent years, we see many contemporary business books that espouse the same idea that Gilbert had. See the Purpose-Driven Economy by 2016 ISPI keynote speaker Allen Hurst. Look at Influencer, one of whose co-authors was 2014 ISPI keynote speaker David Maxwell.
As time passes, simplistic misinterpretation of Gilbert’s writing is common. Judging his masterpiece by its cover, the idea of “engineering human performance” conjures up the image of a pointy-headed solitary figure emerging from his cubicle with a finely-honed solution to underperformance. But for most savvy managers, the idea that you can engineer performance of humans through monolithic analysis is specious or misguided. It certainly flies in the face of how the world of work has evolved, where collaboration and adaptability rule the day. Such an approach puts too much weight on the process side of things and not enough on the human side of things.
This misinterpretation is not Gilbert’s fault. Empathy for the human condition is the subtext of all of Gilbert’s writing. We see poetry; we hear about love; we learn how people are different from animals. He is abundantly clear that the manager’s job is to address systemic conditions that impair the ability of individuals and teams to fulfill their stated purpose. Certainly analysis is necessary, but he doesn’t say that the supporting analysis needs to be onerous or without regard for the people in the system.
Any time you spend doing detailed performance analysis is time that problems continue to fester. What if you did a nominal amount of analysis, and then started with a plan to remediate the biggest obvious problem? Through an iterative cycle of retrospection and re-planning, you make early continuous progress toward your vision. Human Competence provides sufficient tools for doing this analysis in a lightweight way.


“Human competence is a function of worthy performance (W), which is a function of the ratio of valuable accomplishments (A) to costly behavior (B).”
-Human Competence (Tribute Edition) page 18
Being a devotee of Gilbert and an evangelist for his Behavior Engineering Model (BEM) can be exhausting. Outside of ISPI conferences or chapter meetings I never hear anyone but myself talk about, much less apply, what Gilbert taught us.
My book, Agile Performance Improvement, is the story of my learning to blend the mindset of HPT with the values and principles of Agile. HPT teaches us the value of front-end analysis, including root cause analysis, so that you can feel more certain that your efforts focus on the right things. Agile provides a set of methods to guide the actual doing of the work, the process of making performance improvement happen.
The way you did work in the 1978 will not really fly in 2018. The Agile Manifesto (and the methods inspired by it) are based on the idea that preplanning large (creative) projects introduces risk, because the moment you plan every task and dependency occurs at a point when you know less about the situation than you will know at any point thereafter. The iterative nature of Agile methods ensures that you regularly revisit and re-plan based on what you have learned (usually every two weeks). The notion that one person with a few spreadsheets can “engineer performance” undermines the idea that in a truly collaborative, empowered and nimble team environment, everyone can contribute to engineering worthy performance.
These are the reasons why Gilbert still matters. I have not read every book in our field, but among the ones I have, Human Competence remains the greatest, even 40 years after its initial publication.


Gilbert, T.F. (1996). Human Competence: Engineering Worthy Performance (Tribute Edition). New York: ISPI.