An Alternative to Agile Coaching
Over the last 6 years the role of Agile Coach has emerged in the IT workforce. I’ve worked as one for the last 5 years and most of my work has been done with Suncorp, a large 16000+ person Australian insurance and banking company. Suncorp is known as a leader in Agile adoption and as an example of how Agile can transform an organisation and help it achieve outstanding results.
When Suncorp called me in, back in 2007, to help them transform their entire IT services organisation of around 2000 personnel to an Agile way of working, it was a daunting challenge. There was no structured curriculum available anywhere in the world that covered all roles and maturity levels. There was no Agile training available for Leaders, Project Managers or Change Managers. We had to build it from scratch and that’s how the Agile Academy was born. There were no experienced qualified Agile Project Managers and the Agile maturity of most teams was very low. Furthermore it was hard to find Agile experts with experience and knowledge at a management level. We therefore had to spread the few Agile experts we found over many projects as Agile Coaches. The Coaches reported to a central Coaching Manager and had a dotted line reporting in to the Manager of the project or team they were coaching. This Coaching model suited the purpose, worked well and we slowly but steadily built up the capability of the entire organisation over the last 5 years. Coaching did serve its purpose and was the only option at that time.
But now extensive, high quality Agile training is available at all levels and for all roles, there are quite a few Agile experts in the industry and the overall of level of Agile maturity is much higher.
So is there a better and faster way of building the Agile capability of an organisation? Is there another way that can complement coaching?
It’s now time, in true Agile fashion, to look critically and constructively at the coaching model and see if there if there is an alternative way of building Agile capability.
Original objectives of the Coaching role
The benefits of Agile are now well established and I don’t intend to go into it those here.. Organisations have come to realise that if they want to increase productivity and deliver value faster, cheaper and better, Agile is a proven way to go. Its core principles of teamwork and collaboration, iterative delivery, flexibility to change, focusing on business value and continuous improvement transform teams and organisations and deliver benefits from day one in terms of faster time to market, reduced cost, increased morale and improved quality.
The key objective of the Agile Coach role was to help with this transformation and lift the Agile capability of every member of the team. The success criteria, defined by the change team and the sponsor, at the start of the transformation journey was the sunset of the coaching role.
Challenges of the Agile coaching role
Looking back, I have found that the Agile Coaching role faces a number of challenges and is therefore suboptimal in its effectiveness.
Lack of skin in the game
Agile Coaches don’t have any delivery responsibility and hence no skin in the game. Their commitment to the success of the project and its deliverable is often not perceived as they tend to be focused on the process only and not the outcomes. Because of the lack of skin in the game a number of Agile Coaches didn’t take a pragmatic approach and sometimes alienated themselves from the team and the outcomes. If the Coaches are spread over too many projects, or are only engaged for short intermittent periods, they are sometimes seen to use the ‘Seagull’ approach where they just fly in, squawk a lot, crap all over the project and fly away. This approach actually increases the resistance to change and defeats the purpose of increasing the Agile capability of the entire team and achieving transformational change.
Perceived as an unnecessary overhead
Projects often say they want an Agile Coach to help them but when asked to pay for the Coach from their project budget they cringe and back out. This is often because they have not factored in the Agile Coach role in their original cost estimate or they see the role as an additional overhead on the project. The value of an Agile Coach is hard to quantify and measure and hence the reluctance to pay for one.
Lack of authority or a clear role in the team
Agile Coaches have very little authority in the team and often report to the Project Manager or Team Leader. They offer suggestions and advice but have no authority to make sure that the team follow the recipe in order to learn the correct approach.
This is why very often we see teams adopting Agile ‘partly’ or selectively. More often than not they go against the advice of the Coach and all the Coach can do is whinge a little and finally just walk away. Furthermore it’s very hard for the Coach to hold any individual in the team accountable for not following a proposed approach or practice. The biggest culprits are sometimes the Leaders or Project Managers, to whom the Coaches report. It’s hard to hold anyone accountable if they are not given suitable responsibility to match.
‘Coaching’ alone is not the ideal model
Unlike Consulting, Coaching involves not giving answers, but asking the right questions so that the person or team can come up with the appropriate answer. In Agile immature teams and teams that are inherently resistant to the change, the adoption of the coaching form of capability building alone, is not appropriate and does not work well. One needs a more consultative, instructive and training oriented approach with teams who are taking their first Agile steps. An Agile Coach has to train, consult and coach based on the situation. However the percentage of true coaching activities is very small, compared to the consulting and training components. When an Agile Coach takes a consulting approach, teams and the Project Managers could get a bit defensive and resistant, as what they asked for is an ‘Agile Coach’ and not a consultant or trainer.
Calibre of Coaches
At a recent Agile Australia conference there was a panel of Agile Coaches talking on the subject and one of the questions from the audience was ‘How many of you have had any formal coaching training’? Only 1 of the 5 had some form of coaching training.
In my years as an Agile Coach I have seen some great Agile practitioners with indepth Agile knowledge and a breath of Agile experience, but only a few who have been trained as an Agile practitioner and as a Coach.
The main criterion for becoming an Agile Coach has been ones knowledge and experience of Agile. The ability to train or coach is seldom taken into account. More recently, with the demand for Agile Coaches increasing as more and more companies try to go Agile, a large number of Agile practitioners, with little or no formal coaching training, certification (International Coaching Federation) or experience, label themselves as “Coaches”.
In the words of Bruce Weir, an Executive General Manager, at Suncorp, “Give me someone who can lead the team in an Agile manner and lift their Agile capability at the same time. I want someone who has skin in the game and can be held accountable with clear measurable goals”. Sandra Arps, the head of a PMO in Suncorp says “I need experienced capable Agile Project and Change Managers. Not Coaches who float in and out and whose value is hard to measure”.
I think the Agile Coaching model did deliver value at a time when the overall maturity was very low and almost no structured training existed, when it was impossible to find experienced Agile Leaders, PM’s or IM’s (Iteration Managers/Scrum Masters). But times have changed and experienced Agile practitioners are now more easily available. I feel that the Agile Coaching model has its drawbacks and is not necessarily the only model or the best model for increasing Agile capability.
So what’s the alternative?
I would hate to ‘do a seagull’ with this article so I hereby put forward an alternative suggestion or solution for the challenge of lifting Agile capability and transforming an organisation to an Agile way of working.
If you look at a Coach of a sports team, they are responsible for the success of the team and not just their capability. If you look at the Lean process or at Six Sigma, the focus is on leaders as teachers and they seldom use external Coaches that don’t have delivery responsibility. The leaders do the coaching as one of their tasks.
We don’t have Java or Prince2 Coaches or Coaches for any of the other key capabilities in the IT industry, so can we learn from their models? Most other capabilities have the practitioners train the other team members and the team while on the job.
I believe an alternative model that could work very well is having trained and experienced Agile Practitioner Managers (APMs) build the capability. These could be Team Leaders, Project Managers, Delivery Managers, Change Managers or Iteration managers (read Scrum Master), in charge of and running the Agile teams and this would deliver the best outcome for the project and the best capability uplift for the whole team.
The APM model avoids all the pitfalls of the current coaching model as listed above.
APMs would have definite skin in the game and be driven to deliver successful projects and outcomes.
There would be a budget to fund them, as they would not be seen as an overhead. Every project has to have a Project Manager and/or an Iteration Manager.
They would have both the responsibility and the authority to push and cajole resistant teams in the right direction, build their capability and be measured and held accountable for it.
The APMs would be best positioned to offer a blended model of learning which combines leadership, consultancy, coaching, mentoring and training where necessary.
I also believe the core capability of these APMs would be the fact that they are first and foremost good leaders and managers and then agile practitioners, as opposed to being an Agile practitioner with marginal real world experience in leadership.
The APM model would address all the pitfalls of the Coaches model and be able to deliver more bang for buck.
Agile has matured and come a long way over the years but it is now time to take it to a new level and build in ways to accelerate the learning curve. What is Agile 2.0 or 3.0? Whatever the answer to this question is, whether it be ‘Continuous Delivery’, ‘Experience Design’, or ‘Product Management’, it is clear that the way capability is built and organisations transformed will also need to adapt to keep up with the changing Agile maturity landscape.
How could you make it happen?
The key challenge is finding or developing APMs. As the Agile way of working is relatively new it’s very difficult to find experienced or even trained Agile Leaders, Practitioner Managers such as Agile Project Managers, Agile Change Managers or even Agile Iteration Managers/Scrum Masters. If an organisation wants to embark on an Agile transformation journey, where do they start? How do they find or develop these APMs to support their Agile transformation program?
Every organisation has a number of Project Managers, Change Managers or Delivery Managers, whether they are contract or in-house staff, who already know the organisation and have the experience in their core management practices. A good place to start would be to corral and intensively train these people to be APMs. Then embed or assign them to the critical projects that are selected for the Agile journey and have one experienced Agile Super Coach (an Agile trainer, or senior Agile Coach) on hand to help them on their Agile journey. This Agile Super Coach can support multiple APMs and provide the hands-on, onthe- job, training they would need to employ the practices they learnt in the classroom. These APMs could be grouped together in a central team or virtual team, and assigned out to projects on request. By being in one central team the APMs could increase their Agile capability and share and learn from each others in a more efficient manner. The projects would be willing to fund these APMs because they would have budgeted for a PM or IM anyway and won’t see it as an overhead.
It is also a good idea to train all the organisational Leaders (Line and Executive Managers) as well as the APMs so that they can support and steer the APMs in an Agile manner.
In summary, I feel that a good blended learning programme, both online, class room based and on-the-job, for Agile Practitioner Managers, followed by the spread of these managers across projects, with support from one Agile Super Coach, would be a very effective model for rolling out Agile as a way of working in an organisation. This APM model could be blended with a normal Coaching model in geographic areas where Agile training and Agile expertise is still not easily available.
The APM model can be more effective in speeding up the Agile maturity of an organisation and can be used in conjunction with the Coaching model if needed. It’s time for Leaders and the Agile Practitioner Managers to coach, as any Leader normally does as part of their job, and grow the Agile capability of the team and the organisation as a whole.
About the Author
Phil Abernathy is an inspiring Agile/Lean Leadership speaker, trainer, coach and consultant. His mission is to make every work place a happy place. His passion is making a material difference to the top and bottom lines of companies by substantially lifting the capability of their Leaders. With over 30 years of experience in blue chip companies all over the world, Phil is the owner and founder of Purple Candor, an Australian company focused on enabling leadership excellence.
Are you describing agile or traditional projects?
In your world agile seems to be closer to traditional models of IT development with project managers, team leaders etc.
Is Suncorp really doing agile?
The organisation you described does not look agile to me:
* project managers
* iteration managers
* agile coach reporting to project manager
* Agile Coaches have very little authority
I could go on and on, so all this article did for me is raised the question if Suncorp knows anything about agile at all?
Agile pasted on top of traditional management culture
Successful coaching has nothing to do with "having skin in the game". That's a myth. A coach cares about the people, and about integrity. One can be utterly indifferent to a product, or to a company's financial success and yet still be a great coach—still care about the people and about creating right relationships. In fact, I'd argue that a coach (or "Agile manager") that cared too much about business success would be a hindrance to change and improvement.
And just so I don't "do a seagull" :) I'll offer my own vison on the future of embedded Agilists: bit.ly/I7YCTe. ...and some Thoughts on Coaching: bit.ly/Iq6ZJI
Prince 2 coaches exist as well. They're called consultants. An Agile Coach is a consultant who specializes in Agile practices and many of them do have coaching backgrounds.
Re: Are you describing agile or traditional projects?
Moreover I never saw the reason for Agile coaches being the replacement of 'management and control'.
I was describing Agile Projects and the role of the Agile coaches was to lift the Agile capability of the team and not to replace management and control.
Re: Is Suncorp really doing agile?
Suncorp is practising Agile in a very principled yet pragmatic way. Suncorp is a very large corporation and the projects being run Agile are often in the $5-10 million range with one even in the $50million range. To think that these projects can be run without Program/Project Managers or Scrum Masters ( Iteration Managers) is not realistic.
You do imply that if a project has a Project Manager or Iteration Managers (read Scrum Masters) they are not Agile and that is something that surprises me.
As for the Agile Coach reporting to the Project Manager, well that does happen sometimes and as I did point out in the article, its not ideal. I would prefer for the Agile Coach to report to the Sponsor or even the Steering Committee.
Agile coaches are 'coaches' and in most organisations that I have worked in they do not have any authority over the team. Could sometimes help if they did have some authority, but then would they be coaches in the true sense of the word?
Re: Agile pasted on top of traditional management culture
I totally disagree with your point that a coach can be a 'great coach' and be indifferent to the product, outcome or financial success of the company. I do agree that a coach cares about people and by default if they care about people they have to care about the results of the company as without that the people would not have a job!!
If this approach is working well enough for Suncorp, great, best of luck to you. To me, it smacks of dysfunction, but I realize a large organization has to start somewhere. I'm not naive enough to think every organization can only benefit from 'Agile' if they 'live the mindset' and other similar blanket statements coming out of the Agile community. I'm working for one such company now and it will take years to move away from their current system of managing work.
I'd also suggest based on your references to 'skin in the game' and 'coaches cannot be great if they are indifferent to the product' shows a lack of understanding about coaching in general. A good coach knows you cannot force change into any system, and the minute you have product expertise and have 'skin in the game', you've robbed your client of learning and have created a dependancy on the coach.
Ken Power is doing some great work on scaling approaches ( systemagility.com/ ) and Scott Ambler is too. It would be interesting to see how this model works out in comparison.
I did suspect this article would ruffle a few 'coach' feathers. Please dont take it as an affront to Agile coaches...I'm one too and being constructively critical, of self as well, is part of the learning process. The article carefully outlines my experiences and proposes the APM model as a complementary model to Agile coaching.
As for whether coaches should care about the outcome of projects or just the people aspect, lets just agree to disagree. I think a good coach should consider both aspects and I respect your view otherwise.
Experience Report? Theory Piece? Market Research
As I was reading your article I could not work out whether this was an experience report (Suncorp), a theory piece (this is what I think we should be doing or whether this is what you have done) or Market Research (Are you generalising your experience at suncorp or is this based on some poll or market research).
The Agile Coach role grew out of the XP coach role. A role that has been around since before 2000. Many of the XP coaches found they liked the role and evolved into Agile coaches to support team transitioning to Scrum. I agree that the role probably became popular in the mid 2000s but people have been doing it for a long while. Sadly when organisations embarked on Agile transformations they often did not do their research and ended up chosing people without experience. Qualifications as we know are particularly suspect in this area. That said, many coaches had close relations with the IAF (Institute of Applied Facilitators?)
Agile (we really mean Scrum) and XP are processes and what you want are people with experience applying them so that they know where the bodies are buried... which corners can be cut, and which should not (like shutting down the PMO department). Often coaching refers to personal coaching and qualifications are probably desirable.
From your intro it seems that all of your experience of coaching is at suncorp. Others think that Suncorp does not sound particularly Agile. Before you try changing Agile coaching, you may want to try coaching in an Agile environment. You may discover you need a different skill set and do not need to justify your existence too hard.
Thank you for an interesting read. It would be good to see the experience report (at Agile2013?).
Agile principles do not require the role of a coach. This is a construct we have created. In fact Agile's focus on a cross-functional teams should discourage the 'agile coach' role as much as a 'tester' role. But this rarely happens. Add to this an Agile coach may be helping on various projects and be doing extreme task switching. This also goes against some fundamental agile principles.
I'd like to ask the reverse question, 'what are the drawbacks of having an 'embedded agilist' versus an 'agile coach'? I can't think of one.
My experience using a similar approach
I was brought into an R&D department of roughly 80 people to provide them with the agile knowledge that they needed to change to an agile way of working. When I arrived the organisation had been split into nine Delivery Teams, each with an Iteration Manager. The organisation was still operating with a Traditional / Waterfall development approach. There was a core management team with over-arching responsibility for delivery, support and people.
We had a Transition Team (Scrum team) that was made up of a cross functional group of people from Delivery Teams. I was the Scrum Master for the Transition Team and the head of the department was the Product Owner. This team worked through a backlog of organisational impediments and did a great job of it.
My initial responsibility was to educate the staff on agile, Scrum and agile technical practices, with a strong focus on the Transition Team and Iteration Managers. This was primarily done via class room type training and some coaching. One of the Iteration Managers stood out as ‘getting’ the agile message and being good with people hence she joined me in the next step.
As we progressed closer to the cut over from Traditional development to Scrum our focus turned to coaching the Iteration Managers, with them split between myself and my off-sider. The intention was that the coaching we gave to the Iteration Managers would flow through to their teams through them. The Iteration Manager met regularly at a community of practice to discuss what they had learned. This approach continues to this day with all teams sprinting, two of us coaching the Iteration Managers and the Iteration Managers coaching their teams.
Depending on how each Delivery Team is progress we apply our focus more too specific teams; however it has been rare for us to full time with any particular.
While not a contributor on any team my yearly appraisal was tied to over organisation success and especially success of the transition; so I had skin in the game.
This approach has allowed us to educate and coach the organisation, reaching out to about 80 people. We have delivered multiple releases using Scrum and continue to improve. This approach is drawing us ever closer to the point where I am no longer needed, and I like to call that success.
Coach or trainer?
as far as I can see, you understand the role of "coach” more in the way that I see the role of "trainer". For me, the "coach" approach should not embrace engagement in giving advice or playing any kind of a role in the team (as you said "being the skin”). The role of "coach" is to support the team in achieving own solutions, asking questions that cause the team to think things over, end so on. Also the role of Scrum Master is different because SM needs not only to coach the team but also to be the facilitator (removing obstacles).
We don’t need to make up the new roles as „An Alternative to Agile Coaching”. The coach is the coach and other roles were discribed and are well-know, see here:
Maybe your approach differs because we understand the term of "coach" so differently? What do you think?
"Agile Project Managers"
Agile Project Manager is an oxymoron. Like Military Intelligence and Accurate Estimates. #agile #scrum
Re: Coach or trainer?
Phillip states, "Coaching involves not giving answers, but asking the right questions so that the person or team can come up with the appropriate answer."
So his definition of a coach is similar to your definition (The role of "coach" is to support the team in achieving own solutions, asking questions that cause the team to think things over, end so on.)
I applaud Phillip and fully support the idea of looking for alternative approaches to the "traditional" coaching role. One core foundation of my coaching (along with agile principles and values) is to be reality based. The reality in many organizations is fully expressed in Phillips article (see pitfalls above). I do believe the APM model also has an advantage while scaling an agile practice across large organizations. Using the APM model to facilitate maturity within an organization is a brilliant option. It’s an option (not "the only" option) as all implementation/transformations and organizational cultures are different.
Coaches are low value
I don't understand the details of Suncorp, but in general, I think agile coaches are a tremendous waste of time.
They don't architect, they don't develop, they don't manage, what do they do?
Cheer and suck money out of budgets?
Most companies would be far better served utilizing the funds to hire better developers and architects, which make meaningful long and short term contributions, which result in better productivity.
Getting (even first rate) coaches to turd polish some 3rd rate team is a losing proposition.
Agile coaches don't understand the technicals enough to manage, architect, or develop, so they really are useless.
If they think they are not, then they can try to coach football, or voice, or concert piano, and see how far they can get on bravado alone without the skills and background to make said coach worthwhile in said domain.
Brandon Holt, Preston Briggs, Luis Ceze, Mark Oskin May 21, 2015
Kai Kreuzer, Olaf Weinmann May 21, 2015