Architecture Workshop with Legacy Systems

This is an agenda I used recently in a 2-day workshop on a future architecture with lots of existing system and Code legacy („brownfield project“):

  1. What is the Business Vision?
  2. What is the current Architecture?
  3. What drive a future Architecture?
  4. How could an ideal system look like?
  5. How do we get there? (Buy vs Build)

We started with the business vision and the management priorities and the current architecture situation. From there we elaborated an ideal system with potential strategies to get there.

Send us an email if you want to learn more or hire us.

Business Vision

Business Vision and Management Priorities
  • From the business vision, derive influencers on the architecture
    • What makes us cringe?
    • What makes us no worry?
  • Collect Functional and Nonfunctional items

Current Architecture

Working on the Current Architecture
  • Identify
    • pain points
    • user activities
    • top 5 use cases that drive the architecture
  • Create model(s) of the architecture to share understanding
  • What is slowing us down?
  • How long are the feedback cycles of work in the teams?
  • How can we enable Continuous Delivery?

Future System

What is the ideal system? How does it look like?
Where do we predict changes to happen in the future?

  • Work in smaller groups and capture ideas
  • Capture different approaches and compare
  • Data flow, security, user interaction
  • Use Event Storming to find user interaction points

Gather and merge ideas into architecture model(s)

  • What is the ideal architecture?
  • Use C4 Context and Component Diagrams
  • How can we minimize blast radius of issues and incidents? –> How can we create autonomy in teams?
  • Collect Glossary with common terms for shared understanding

How to get there?

  • What next?
    • Collect baby steps to get there
    • Collect big bang to get there
    • Buy or Build? Sub systems
  • Brainstorm activities and trade offs
  • Create a backlog of work

Pain Points

There are lots of points we address in the training or by running the workshop with your real operational systems architecture, like:

  • How do we know “Enough modeling is enough”?
  • When should we take which decision?
  • When do we start to digitize the models? (Why should we?)
  • Which notation makes sense?
  • What is a great architecture?

Tools we ❤

Here some tools we use in this Architecture workshop

  • C4
  • Clean Architecture
  • Wardley Mapping
  • Impact Mapping
  • Domain Driven Design
  • Event Storming
  • User Story Mapping
  • Ports and Adapters or also known as Hexagonal Architecture (by Alistair Cockburn)
  • The Universal Architecture (Concept by JB Rainsberger)
  • Pen, Paper, Post-Its

What does an Architect do? (Yes, the role)

From Business Vision to Leading the way.

Send us an email if you want to learn more or hire us.

Peter Gfader on Twitter Send Peter a tweet

5 Ways How Enterprises Can Successfully Act Like a Startup – My recommendations

Debo Olaosebikan @dolaoseb describes 5 Core Principles in his blog post “Big Winners Start Small

  1. Understand what users desperately want
  2. Test your understanding with cheap experiments
  3. Measure what’s important
  4. Build as little as possible
  5. Move fast and iterate

And a Roadmap to Go from Big to Small

  1. Create Small Projects
  2. Form Small Teams
  3. Differentiate between Product Management and Project Management
  4. Leverage Existing Technologies and Tools
  5. Nurture Entrepreneurial Teams

My comments

I like the roadmap, and in my experience for this change to work, you need to change the organizational constraints around “Projects” which include funding of them.
For me, that means I would start with “0.a Remove Legacy Management Practices” and then “0.b Differentiate between Product Management and Project Management

At the end it would look like the following.

Roadmap

  1. Work with Management: What is hindering the value stream?
    1. What legacy management practice is hindering your employees?
    2. Where do we create value? Where are we just busy?
  2. Work with Management: How does money flow?
    1. What do we incentivize?
    2. Where is our focus and what do we measure?
      Internal metrics or external customer-centric metrics?
    3. What are Projects? What are Products?
  3. Work with 1 Team: How can we make it work in the small?
    1. Create 1 Small Team with external customer access
    2. Establish Trust in your organization
      I wrote a german article about: Why the management doesn’t trust the development teams.
  4. Work with multiple Teams: How can we scale this approach?
    1. What is the common goal and vision?
    2. What is the network of supporting people for the teams?
    3. Nurture Entrepreneurial Innovative Teams

And remember:

Kanban and Scrum don’t matter.
Your process doesn’t matter.

What matters is that you have a process to improve your process.

Sign up here to get notified of new articles.

If 95% of all Startups Fail, Why Should Enterprises Behave Like a Startup?

Great question! via Roman

I think “enterprise projects” should behave more like a startup!
By behaving like a startup I mean:

  • Reduce risk and uncertainties
  • Validate hypotheses and learn
  • Prototype and get user feedback early
  • Find paying customers: That means in an enterprise –> deliver usable features that people ❤ (love)

Most of the big projects do these things in some sort.


But….

What they don’t do usually is:

  • Pivot: 
    “Hey this is not what the user wants, maybe we change something”

  • Kill it 
    “We invested 10 million in this, let’s add 1 more million then we are ready” (The sunk cost fallacy)

“95% of startups fail”
Ok…. In my experience in an enterprise context, people tend to ride the dead horse too long. 



Sign up here to get notified of new articles.


Original blog post was posted on https://business-agility.tumblr.com/ on June 2016 here
https://business-agility.tumblr.com/post/132714222320/if-95-of-all-startups-fail-why-would-enterprises

Strategies to Ride a Dead Horse

If this website goes down again, I have a backup here.
All the below was copied from http://www.dbrmfg.co.nz/Next%20Step%20Dead%20Horse%20Strategies.htm


Dakota tribal wisdom says that when you discover you’re riding a dead horse, the best strategy is to dismount.  However in business we often try other strategies with dead horses, including the following;

  • Buy a stronger whip.
  • Change riders.
  • Threaten the horse with termination.
  • Say things like, “This is the way we have always ridden this horse.”
  • Appoint a committee to study the horse.
  • Arrange to visit other sites to see how they ride dead horses.
  • Lower the standards so that dead horses can be included.
  • Appoint a tiger team to revive the dead horse.
  • Ride the dead horse “outside the box.”
  • Buy a commercial off-the-shelf dead horse.
  • Create a training session to increase our riding ability.
  • Reclassify the dead horse as “living-impaired.”
  • Compare the state of dead horses in today’s environment.
  • Change the autopsy report to declare that “This horse is not dead.”
  • Kill all the other horses, so this one will look the same.
  • Name the dead horse “Paradigm Shift” and keep riding it.
  • Ride the dead horse “smarter” not harder.
  • Hire outside contractors to ride the dead horse.
  • Harness several dead horses together for increased speed.
  • Do a time management study to see if the lighter riders would improve productivity.
  • Declare that “No horse is too dead to beat.”
  • Call the dead horse a “joint venture” and let others ride it.
  • Provide additional funding to increase the horse’s performance.
  • Do a cost analysis study to see if contractors can ride it cheaper.
  • Purchase an aftermarket product to make dead horses run faster.
  • Declare the horse is “better, faster, and cheaper” dead.
  • Form a quality circle to find uses for dead horses.
  • Declare that “This horse was procured with cost as an independent variable.”
  • Get the horse a Web site.
  • Promote the horse to a supervisory position.

Time for a question.  Did any of these seem vaguely familiar to you?  A few I suspect.  So, now that you have read about all the things that you shouldn’t do and already know, please go to the beginning of this website and check out all the things that you haven’t read about that you should do and don’t know!

The dead horse strategies were reproduced with the permission of Bill Dettmer.  From Appendix F, Strategic Navigation.  ASQ Press (2003).