Our colleague Ignacio Elorriaga, Agile Coach and Scrum Master of dareCode, through this article makes a comparison between the Agile Frameworks: SAFe, Less and Nexus.
If you’ve come this far, it may be because you’re thinking of scaling up one or more projects you have within an area of your company.
In that case the first thing I recommend is that you ask yourself, is it really necessary to scale those projects and “unify” them?
You must take into account that the use of these Frameworks, described below, raises the cost and generates some overload of meetings that will consume for the correct synchronization of the teams.
As a preview, you may ask yourself, should we change the structure of our organization with one of these implementations?
If you have to implement it, think that this goes beyond assigning new labels to old roles and ceremonies.
Having said that, let’s look at the differences and similarities in broad strokes (we won’t describe every element because that would be endless) between the following agile frameworks for scaling projects. The main proposals are SAFe 5.0, Nexus, LeSS.
Table of Contents
What does each Agile Framework mean? How are they defined?
As each is promoted by a different body they have different nomenclatures for similar concepts, for example, Scrum Sprint in SAFe is called “Iteration”, and in Nexus & LeSS “Sprint”, and a long etcetera.
SAFe: Scaled Agile Framework.
Created by Dean Leffingwell (official website) is defined as: ‘SAFe for Lean Enterprises is a knowledge base, integrated principles, practices and proven competencies for Lean, Agile and DevOps’.
Powered by Ken Schwaber, Scrum creator, and the Scrum.org team (Nexus official website). It is oriented to the development of software and support of those products that are scalable.
Defined as a ‘framework consisting of roles, events, artifacts, and techniques that go together and intertwine the work of approximately 3 to 9 Scrum teams working on a single product portfolio to create an integrated increment that meets a goal’.
LeSS: Large-Scale Scrum
Created by Craig Larman and Bas Vodde (LeSS official site). This is the lightest configuration when coordinating Scrum teams of three.
LeSS wants to establish agility by changing the structure it is in and its organizational policies through the roles imposed for synchronization (avoiding ‘Fake-agile’).
Agilistic Frameworks values and principles
All three are Agilist Frameworks so they are based on Lean and Agile, and therefore on the Agile Manifesto, with its 4 values and 12 principles.
But also, by using Scrum on a daily basis:
- They are based on the three pillars of Scrum: transparency, introspection and adaptation.
- They will also support the Scrum values: Courage, commitment, Focus, Respect, Openness.
However, as they are “extended Scrum”, since they are extended practices to manage several teams where Scrum does not reach, they have their own values.
Except for Nexus which, being 100% based on Scrum, has no values or principles of its own, only those already defined in Scrum.
SAFe's 'Lean-Agile Mindset'
These are the values on which the whole framework is based: Value (above all), supported by: Respect for culture and people, Flow, Innovation, Continuous improvement, and all of this supported by leadership.
And in addition:
- It has 7 competences
- 4 values: Built-in
- Quality, Program Execution, Alignment, Transparency.
- 10 Immutable principles and behind Lean-Agile.
LeSS's 'Lean-Agile Mindset'
In LeSS we also find its own rules:
- 10 rules for the structure in LeSS and 6 in LeSS Huge
- 7 rules for the product in LeSS and 4 in LeSS Huge.
- 11 for Sprint in LeSS and 2 in LeSS Huge.
And it’s also based on a list of principles because according to them the rules are minimalist and do not answer how best to apply LeSS in your specific context.
Methodologies used in SAFe, Nexus and LeSS
As we are talking about scaling, all of them will implement at a micro level (team) some framework to work on their day to day.
|At team level
||Essentially Scrum, but you can also see Scrum XP.
DevOps equipment works in Kanban.
|At teams level
||Kanban is used in the Portfolio and Program & Solution display||
Configurations available in SAFe, Nexus and LeSS
Depending on which level you want to scale your projects, these Frameworks allow us to implement them at different levels, ranging from 2 projects/teams to the whole company.
We must indicate beforehand that in SAFe the teams working together are integrated in a “train”, formally called Agile Release Train (ART).
On the other hand, in LeSS Huge, every three teams are integrated into one Area.
|Essential (5 to 12 teams in an ART).
Large, one more level of abstraction to align several ARTs together.
Portfolio, one more level of abstraction to reach the whole company.
|One single configuration valid for managing between 3 to 9 “Scrum Teams”.||LeSS covers between 2 and 8 teams.
LeSS Huge, from 8 projects.
SAFe has a ‘Portfolio‘ configuration and it is the only one of the three that takes into account the organization, that is why it can be scaled to the whole company.
On the contrary, Nexus and LeSS only take into account the organization of the product in different teams, dividing them into areas at most (LeSS Huge) and here it can be extended to dozens of teams, but there will not be a synchronization, predefined within the Framework, between areas.
We can then point out that SAFe focuses on the whole organization and LeSS and Nexus pay more attention to the products.
Events in each Agile Framework
Because they are based on Scrum, they share team events even if they have different durations and names.
Similarly the three establish that the Sprints (Iteration for SAFe) must begin and end at the same time for all teams involved and is the event that contains the rest, being common to all teams.
That way you get one more level of synchronization within the computers.
The duration of each Sprint is defined by the Framework itself: in SAFe they say it should be 2 weeks (recommended, but they also accept between 1 and 4 weeks), in Nexus as it is based on the Scrum Guide up to 4 weeks, and in LeSS between 1 and 4 weeks.
|Team level||Sprint Planning
Daily Stand-up (DSU)
|Sprint Planning “2”
Product Backlog Refinement
|Teams level||N/A||Scrum-of-Scrums (SoS)
Inspect and Adapt
Program Increment (PI)
|Nexus Sprint Planning
Nexus Sprint Review
Nexus Sprnt Retrospective
Nexus Daily Scrum
|Sprint Planning “1”
Overall Product Backlog Refinement
When describing the events it is worth noting that for SAFe the main event for everyone is the “PI Planning” since it is “where the magic of the ART is created” and that without this event no SAFe is being done (it is a planning every 8-12 weeks where all the ART is gathered in one place to decide what to do in the next PI).
On the other hand, the rest of the events in all the Frameworks will have as participants the representatives of each team or the specific roles, as for example, the SoS will go only the SM of each team, or the Nexus Daily Scrum the representatives of each Scrum Team.
Main roles in each Agilist Framework
Everyone has, in essence, the same roles at team level but adds new ones with different obligations depending on the position they hold within the configuration.
|Team level||Product Owner (PO)
Scrum Master (SM)
Agile Team (3 a 9 personas)
Development Team (3 a 9 people)
Feature Team (3 a 9 people)
|Teams level||Product Manager
Release Train Engenieer (RTE)
|Nexus Integration Team [NIT] (its like an Scrum team)
||PO (1 for all)
In Less Huge:
- PO Area
- Feature Team Area
|Portfolio level||EPIC Owners
Lean Portfolio Management (LPM)
Here one may wonder if it is necessary to have an SM / PO for each team (which would be a high cost if we had +10 equipment).
From experience, an SM can manage 2-3 teams (and 3 would seem a lot to me) at the same time (bear in mind that Sprints are shared and therefore all events are at the same time).
So if necessary we could have a PO leading 2 teams with similar characteristics or an SM in two teams, more serious would be to have a “shared resource” within the train.
SAFe, Nexus and LeSS Artifacts
These are the different elements that are used, or generated, and are defined in the guides for each one.
|Team level||Iteration Backlog
Burn-down chart, Cumulative Flow Diagram (CFD)
Definition of Ready
Definition of Done
Definition of Done
|Teams level||Program Board
|Product Backlog (the same for all)
Nexus Sprint Backlog
|Product Backlog (the same for all)
En LeSS Huge:
Area Product Backlog
|Portfolio level||Strategic themes
We have explained very briefly the characteristics of the different scaled Scrum frameworks.
We hope this will give you a little insight into the different approaches currently available.
If you liked this article we will make a more detailed one with concrete problems or situations in which we have applied some of these solutions.
At dareCode we have experts with a long history in scaled Scrum and Agile Coaching.
Discover our Agile Professional Services adapted to companies that want to work with Agile in the right way, whatever their level of maturity in agile methodologies.
Do you want to tell us something?