On June 19th, I sent this email to the CPO of the Scrum Alliance. I have had no reply. In a second article today I will offer some related ideas.

I sent an email to the Chief Product Owner of the Scrum Alliance, Howard Sublett, back on June 19th. He had texted me with sorrow, referring to my postings back around then, If the Scrum Alliance Cared and Double Down: Scrum v Devs.

Since those two articles, I’ve also written two other related articles, Increment? What’s that? and Survey says ….

That latter article gives the results of a Twitter survey I did, asking half a dozen questions of Scrum developers. Perhaps the most striking finding was that over seventy percent of developers working under Scrum reported that no one on their team had any Scrum-related developer training at all!

I do think this situation is deplorable, and I do think that the Scrum Industrial Complex has great responsibility for this lack. And I even commend, lightly, efforts made by Scrum Alliance and Scrum.org to offer certifications, and related training, for developers.

Herewith, my letter, with a few irrelevant bits removed.

To CPO, Scrum Alliance

This is long, but I’ve left most of it intact for context.

Perhaps email will serve as a way to communicate thoughtfully and in complete sentences, compared to phone messages. Which, in my case, are actually spoken into my watch and typed out by Siri or somebody. We should also set up a series of zoom or phone calls, but I’d like to get some common ground first.

Chet has pointed out to me, and I agree, that there are some key points of agreement, and some opportunities / challenges that (we think) the Scrum Alliance (you) must surely see and recognize. Those opportunities / challenges may seem insurmountable, but my point is that in fact they are not.

My main questions to you have to do with your mission, that is, the mission of the Scrum Alliance, and how it will measure itself. Secondarily, if your job mission as provided by the board is different from that, I’d like to understand what I may of what your personal job is, and how that’s measured.

Finally, I’d like to know whether you have a truly personal mission in mind, vis-a-vis the Scrum world, anything that may not be in the board’s purview but that you personally would like to accomplish.

My context for the questions above is the stated purpose on the Scrum Alliance web site:

On a mission to create agile workplaces that are joyful, prosperous, and sustainable.

Now, I’m aware that somewhere in the Scrum Alliance, I think coming mostly from the trainers and coaches, there is this notion that Scrum is about any kind of product. However, I believe that by far the bulk of actual Scrum application is in software development. I’d be interested in any actual numbers you have around that question.

I gather (that’s less than “aware”) that a strongly-held viewpoint among trainers (and others?) is that mostly the Scrum Alliance training should be about using Scrum, independent of the particular domain of the product. Is that true? Does it make sense to you? It makes a certain amount of sense to me, with caveats we’ll talk about.

So I have those questions above, and welcome any replies to them and whatever else comes to mind. Now I’d like to turn to something more like my agenda.

There are about a million Certified ScrumMasters, if I’m not mistaken? Some fraction, I imagine, are not actually ScrumMastering. Do we know how many are? Some fraction are not working on software products. Do we know how many are? Suppose that only a quarter of them are working as ScrumMasters on software products. (I’d guess more, but it’s hard to say. Some of them have been SMs for a long time and might be retired or vice presidents or something by now.)

However many products there are out there that are doing Scrum, or trying to do Scrum, driven at least in part by training from the Scrum Alliance, I’d say that those products, and the people working on those products are the minimum group where the mission above holds:

The Scrum Alliance is committed to making the product development workplaces of its trained and certified customers, joyful, prosperous, and sustainable.

Is that fair to say? If not, what would be fair to say that sounds something like that?

Now, with a purpose like that, I’d hope that the Scrum Alliance would have an ongoing effort to be aware of the extent to which the product development workplaces of its trained and certified customers are, or are not, joyful, prosperous, and sustainable.

Does it make sense that Scrum Alliance would want to know that? Do you know it? What are you doing so that you know more and more about that?

Now here’s where my agenda starts:

I am on the receiving end of direct and indirect information about “Scrum” software development where the developers are not joyous, and/or not prosperous, and/of the development is not sustainable. And there are lots of them.

I hear about teams who cannot successfully deliver an Increment at the end of every–or any–Sprint. As you know, that is an essential component of Scrum. As you probably know, it is an element where teams very often fall short.

There is a key reason why teams cannot deliver the increment, and it is actually true for most product domains, though software is the only domain where I think my colleagues and I have some answers. But this is true almost everywhere:

The usual process for any product development is focused on a long cycle of planning, designing, building out infrastructure, finally building the product and then testing it. (Waterfall. It’s the dominant old-school way of building anything, be it a house, a car, or a software product.)

The above are domain-related skills. Your trainers and others who say that’s outside the scope of Scrum are correct, but the fact remains: in most domains, certainly in software development, to build the product incrementally requires new skills, skills that are not present in the engineering curricula, not present in the pre-Scrum work environment.

Scrum teams cannot build the Increment without having special domain skills, outside Scrum. And they do not come out of the box with those skills: they are new, and different, from conventional domain product development skills.

This truth is recognized in the Developer Learning Objectives, at least the fraction that I have seen. Those LOs do refer to important skills like dealing with legacy code, like TDD, like refactoring. And the advanced LOs, the fraction that I’ve seen, do actually use Bloom Taxonomy words that imply that the candidate must really have the skill, not just the ability to talk about it. Words, I forget the specific words, but things like “demonstrate” and “show” and “solve”.

So, while I’m sure my colleagues and I might quibble a bit with the new LOs, they appear to me to be pretty good. I actually said that in one of my screeds, but perhaps you didn’t notice among the many other words.

But here’s my concern: there are surely at least a million programmers, and I think many more, who are working under Scrum. And almost none of them have the necessary programming skills to deliver the Increment. (My recollection from a while back was that your own developers couldn’t deliver an increment, and your main product is a web site, which is about as incremental a thing as exists in software. I don’t know if they’re doing it now. Are they?)

So I think we have a shared mission here: we all want those million or five million developers to be working in joyous, prosperous, sustainable situations. I feel sure that you want that, and I know that the folks on my team over here do.

The problem is, how the hell can we do it?

One constraint, I believe, is that we cannot do it with a CSD program based around the current training model. We simply do not have the capacity to train a million programmers, and based on the success of the program so far, we know that there is a great reluctance to train all of a team’s developers, even at a low low price of maybe $500 a day.

You may be more optimistic. I think the ship has sailed: we cannot reach all the developers with CSD training.


We can reach them by other means. Here’s an expanded set of ideas, including my main two, which I’ll emphasize in bold.

  1. Continue, or begin, to assess joy, prosperity, and sustainability in Scrum product developments, surveying not just ScrumMasters or POs, but all team members.
  2. Bear down, in as many communications as possible, on the importance of the Increment to doing Scrum successfully. Be clear that it is difficult, and requires domain skills, not just Scrum skills.
  3. Continue to skim what cream is available with the CSD training, recognizing that it’s not possible to reach more than a fraction, but assessing whether that fraction is more joyous etc than they were before training.
  4. Provide the ability to earn CSD ratings entirely via SEUs. Specify exactly what videos, what books, what web sites, what trainings constitute CSD-earning applicable SEUs. Use this specification to direct the 95% to forms of learning whose cost in time and money they can afford, even on their own, not company-supported.
  5. Get behind independent providers of training videos, blogs, web sites, and books, cultivating the ones who are producing material that truly equips the developers to deliver the Increment. “Get behind” means at least an outreach, as well as directing Scrum Alliance developers to their materials. It might also mean negotiating discounts for people sent by the Scrum Alliance. It might even mean that the Scrum Alliance would kick in part of the fees for the people it sends.

Please join me in noticing that these ideas don’t take away from anything the Scrum Alliance is doing. Instead they add to the impact that the Scrum Alliance has, on individuals and teams that it has as yet been unable to reach.

The incremental cost of doing these things would be small. There might need to be some expenditure to assess joy, prosperity, and sustainability, but since those are the mission, it needs to be done anyway.

There might be some additional administration somewhere in there.

There might be some actual money spent to support creation of new materials, but I believe that could readily be earned back from fees for their use. (Won’t work for books, royalties are too small. Will work for on line training, in many cases.) But financial support is optional: just directing people to the right idea sources is the main idea and you can readily farm that out to people like me.

What’s the big issue between us? I feel increasing urgency to help these people, and since I see what’s necessary as being so small (see above), I am honestly impatient, perhaps only with myself, for being unable to get the importance of this, and the low associated cost, across.

What do you think?


Ron Jeffries

The response: silence, so far.