Can Kanban on its own support Agility?

"Kanban is for repetitive similar tasks" - Agile Coach

If I had to describe Kanban in under 10 words I would say that "Kanban is for visualizing and managing flow". I'd agree that it lends itself nicely for repetitive tasks but can be used for much more than that. 

It sounds like this individual may be putting the practice in a box. If this...do Kanban if that...do Scrum, never shall the two coexist.

Perhaps they should give the Scrum with Kanban guide a read to see how empiricism can be applied to Kanban. 

"Kanban is for repetitive similar tasks" - Agile Coach

The right sentence shall be called: "Kanban can be used for repetitive similar tasks", because Kanban is not a only-agile method. It's a lean method for visualization work and managing flow. Maybe you misunderstood him/her?

The right sentence shall be called: "Kanban can be used for repetitive similar tasks", because Kanban is not a only-agile method. It's a lean method for visualization work and managing flow. Maybe you misunderstood him/her?

@Martin Gradzki, I don't think I misunderstood the person. My point was in reference to the word "repetitive". In the case of either sentences, if Kanban is used for repetitive tasks, then most likely there is no discovery i.e. not much is unknown. For ex: brushing your teeth is repetitive.

However, agile is more suited for tackling complex work, work where there is more unknown than is known. For example drilling till you find water, sometimes you may have to dig at different depths, right? so more is unknown than is known.

If the work is repetitive and not complex, how are we applying the agile mindset?

It made me think, in that specific context, how Kanban on it own supports Agility. If the tasks are repetitive, then they may not require an empirical process control.

Does the repetitive nature of certain tasks reduce the need for Empirical Process Control? How would value, flow, and quality be optimized without evidence, and how would outcomes be made more predictable?

Does the repetitive nature of certain tasks reduce the need for Empirical Process Control? How would value, flow, and quality be optimized without evidence, and how would outcomes be made more predictable?

@Ian Mitchell, I did say "may not require". I wasn't 100% sure, but the assumption I made was that, if the same exact work is repeated over and over again, then at some point, the need for empiricism may reduce as the work becomes more predictable. At that point, I believe we would have best practices and controls to notice any deviations.

However, in a scenario where there is a continuous inspect and adapt at each station i.e. communication and feedback between the business and developers, then Kanban does support Agility.

I guess in the process of this reflection and discussion, I can see how Kanban can be both used for well defined, well understood repetitive tasks, as well as for exploratory work that requires empiricism.

Please feel free to correct me if my assumptions may have missed something, always happy to learn.

In my opinion, the initial statement is incorrect.   Kanban is not for repetitive, similar tasks.   

It can be used in such a case, but that is not what Kanban is for, regardless if an "Agile Coach" is saying that.

My point was in reference to the word "repetitive". In the case of either sentences, if Kanban is used for repetitive tasks, then most likely there is no discovery i.e. not much is unknown. For ex: brushing your teeth is repetitive.

With that argument, you can also say, that Scrum is "repetitive". You always "do" the same: Your Daily has the same time and location, Your planning, review and retro is always at the same time and have the same timebox etc.

In Kanban, it's not that similar. You implement a new method and acting on the results. There is (not like in Scrum) a "pattern" for implementing. You visualize your work how your workflow really is and then visualize your workflow and detect your defects. You are not changing anything, you only visualize, what you are actualing doing and it does't care if it's agile or not.

After that, it's about you, what are you doing with the results. Mostly, you change something and then, you take a look, what's the effect of it (like PDCA). 

Isn't that an inspection and adaptation concept? Yes, and it's not repetitive and also not necessary agile. 

It's agile, if you are able to react on changes of your requriements or changes in the technology. Is that possible with Kanban? It depends...

With that argument, you can also say, that Scrum is "repetitive". You always "do" the same: Your Daily has the same time and location, Your planning, review and retro is always at the same time and have the same timebox etc.

@Martin Gradzki, Consider if you are misunderstanding what I said. I wasn't referring to repetitive events either in Kanban or Scrum.

Kanban originated from manufacturing if my memory holds good. Consider the task at hand is building a fuel efficient car. Building a car is a repetitive task because building cars is a known thing. IF all that is being done is building the same car over and over again, the whole process is repetitive and in my opinion doesn't support agility. However, the same scenario but with the condition that it has to be an extremely fuel efficient car, now, that maybe something that requires a lot of exploratory testing, feedback, trial and error etc. and in such a scenario in my opinion agility is supported.