Frankly, I’m tired of all of the scaling frameworks. They’re mostly driven by three needs:
Creating revenue for the firms creating them;
From a company or organizational perspective, they’re indicative of lazy, buy agile in-a-box, thinking;
And they feed the “certification happy” nature of our community.
And yes, I too am guilty of falling into the above traps.
I think the introduction of Scrum@Scale has ticked me over the edge and inspired me to write this post. That and reading this article by Neil Perkin, which takes a more reflective view to leveraging useful bits from the various scaling frameworks.
This is a bit uncomfortable for me to admit, but I have some confessions to make…
I’m a SAFe SPC;
I’ve attended a 2-day Nexus training;
I plan on attending / co-teaching Scrum @ Scale with Don MacIntyre in September;
I’ve studied (I mean studied!) and contrasted DAD and LeSS;
I’ve actively coached SAFe organizations;
I’ve been leveraging simple scaling techniques (Scrum of Scrums, bits of SAFe) for well over a decade.
So, it’s fair to say that “agile scaling” is in my bones, in my DNA, and that I’m fairly experienced. And it’s incredibly easy for me to meet a larger scale client and begin discussing scaling aspects quite early in our coaching relationship.
Perhaps it’s just me, but I’ve been indoctrinated into the Tuckman Model as THE view or model when it comes to team maturation and overall health.
You all remember Tuckman, don’t you?
He presented the following stages:
- Forming,
- Storming,
- Norming,
- and Performing
- sometimes Adjourning
that teams go through in their evolution to a solidly performing state.
One of the things that it’s influenced in my coaching and leadership style is the predilection to honor the team. That is, once a team is formed and performing, I am loathed to break them up for whatever reason. Even good reasons like the business priorities have changed or there is a desperate need for the skills of one team member in another team. Or even, the team has some dysfunctional relationships brewing.
For years and years, I've been a strong advocate of goal-setting within your agile teams. Ares where I think goals are important include:
- At the daily stand, focusing the conversations towards the teams' goals;
- During sprint planning and at the sprint review, focus towards the sprint goal;
- If you're doing releases, ala SAFe release trains or a similar mechanism, then having a release-level goal is important;
- To me, Definition of Done and Definition of Ready, are goal-oriented. Providing clarity on the teams' constraints;
But...
My friend and colleague Shaun Bradshaw and I were coaching at a client recently. We started to have a conversation about velocity, not directly driven by the clients’ context, but simply in general.
Shaun was focused on velocity as a relevant metric within agile teams to drive conversations between teams and upper management. And I was struggling to “get there”.
Part of his focus was to create visibility around the difference between average velocity and current sprint velocity. Furthermore, the teams and management would be able to see:
- Velocity gaining stability over time (predictability, low variance)
- And increasing over time (short-term burst)
As part of newly formed and/or newly coached agile teams.
Now I really get what he was saying. And I agree that teams in these contexts should be displaying activity and behavior towards those two results.
In the first part of this post, we explored these “rules”:
- Allow Architecture to Emerge
- Treat it Like a Product
- A Picture is Worth…
- Everyone IS an Architect and Everyone OWNS the Architecture
Now, I’d like to continue sharing the final set of four rules for your consideration in your agile architectural travels.
#5) Keep it Simple and Connect to the Customer
This one is quite near and dear to my heart. Why might you ask? Because I really like complexity. I like engineering complex solutions to simple…complex customer problems. And it’s also quite comfortable for me to fall into that over-engineering, gold-plating, doing more than is required mindset.
Why?
Wow, the title sounds quite bombastic, doesn’t it? And I sound quite full of myself, don’t I? Well…perhaps I am.
Nevertheless, I want to go on record with some simple and pragmatic advice for agile organizations and teams when they’re trying to sort out how architecture “fits” in agile contexts.
In no particular order, here are my guidelines:
#1) Allow Architecture to Emerge
I know you’ve heard this before, but it’s really, really important. The difference is moving from traditional architecture, which –
- Performs Big Design Up Front (BDUF), aka Big Bang;
- Develops a “complete” architectural concept before coding begins;
- Approaches construction horizontally, with delayed layer integration
My first piece of advice is this:
DON’T DO IT!!!
Probably the worst possible setup for “team” is spreading them around the country or the world or the universe and expecting them to behave and deliver like a close, cohesive team.
My second bit of advice for those of you that blame it on “Management” and say you don’t have any say in the matter…is:
WRONG, YOU HAVE LOTS TO SAY IN SUSTAINING DISTRIBUTED TEAMS OR CREATING ANOTHER STRATEGY!!!
I hear this situation (excuse) all of the time. An organization has inherited distributed teams yet also wants to move to more agile approaches. They understand that these teams can be less than optimal, but are reluctant to do anything about it. That is but complain about it.
I recently read an article entitled Engineering Culture and Distributed Agile Teams that was published in InfoQ. In it, the editor called out the following strong themes or takeaways:
Key Takeaways
- Team structure reflects in product architecture
- Distributed teams can perform pair programming by using some remote pairing techniques and tools.
- Microservices influence a good distributed team structure
- Increase co-ordination within a team by encouraging T-shaped engineers
- DevOps tools and practices are valuable for Distributed Agile Teams
- Increase efficiency of Continuous Integration and automation testing in distributed teams by using cloud
While the article is titled and implies a focus on culture, it really focuses more on technical tactics or tooling as the key to distributed teams.
One of the topics that I've typically avoided at agile conferences and in discussions with colleagues is software capitalization. Particularly in agile contexts.
The first reason is that it's usually used as an excuse to undermine agile principles and as a means of continuing old practices.
The second is that I find it somewhat...boring. Yes, it's often necessary for many enterprise or larger-scale contexts. But I'd rather the teams focus on innovation, customer value, and deliver than on capitalization.
And the third reason, which is somewhat embarrassing, is that I never had a succinct recommendation for folks on how to handle it in agile contexts. I always felt that I said "it depends" too often.
Excuses
But all of those reactions were excuses. And I do realize that it's a serious topic for many folks. So, imagine my delight when I came across a real-world blog post from Stephanie Davis of Valpak in Tampa Florida that explained how they've been handling it.
And not only that, the post contains some reference post that nicely provides additional capitalization examples and advice.
Wrapping Up
I've shared this post to be my "Go To" reference for capitalization. Thank you, Stephanie, for sharing!
Stay agile my friends,
Bob.
Quite a few years ago, while I was working at iContact, a fellow agile coach approached me with a free offer.
It seemed that he had developed an agile maturity assessment (framework, tool, approach, strategy, etc.) and wanted to try it out somewhere. I’d known him for quite some time and he had some solid agile coaching experience under his belt.
I politely told him “thank you” and that I would “think about it” and quickly closed down the discussion. To be honest, I was initially close-minded to the idea. Here I was an internal technical leader and agile coach in my company. And, to be honest, we were kicking-ass when it came to agile performance and delivery.
But the more I thought about it, the more I started to convince myself that it would be a good idea. Regardless of our performance, we could always get better…couldn’t we?
