The Blueprint

The Unlikely Parallels of DevOps Transformation and Crisis Response Planning

March 26, 2021 Credera
The Blueprint
The Unlikely Parallels of DevOps Transformation and Crisis Response Planning
Chapters
The Blueprint
The Unlikely Parallels of DevOps Transformation and Crisis Response Planning
Mar 26, 2021
Credera

Embarking on a DevOps transformation can be a daunting endeavor, but success is more achievable than leaders might think if they look to an unlikely place. In this episode of The Blueprint, Credera’s monthly podcast on the business challenges facing global leaders today, we’re breaking down the surprising key to launching a successful DevOps initiative: modeling your plans after your crisis response procedures. 

 

Show Notes Transcript

Embarking on a DevOps transformation can be a daunting endeavor, but success is more achievable than leaders might think if they look to an unlikely place. In this episode of The Blueprint, Credera’s monthly podcast on the business challenges facing global leaders today, we’re breaking down the surprising key to launching a successful DevOps initiative: modeling your plans after your crisis response procedures. 

 

Speaker 1:

The blueprint is brought to you by Cordera Cordera unlock extraordinary. Thanks for tuning into the blueprint today. We're going to be taking a deeper dive into dev ops transformation. I am Jeff. Krizek your host with me as usual is Paul. Hey Jeff, how are you doing today? Paul? I've been thinking about, you know, dev ops transformation. I'm cruising, cordero.com . I come across this article. Dev ops transformation is more achievable than you think it's written by our guest today. His name is Simon Simon. Thanks for joining us today.

Speaker 2:

Hi Jeff. Thanks for having me today. Glad to be here.

Speaker 1:

Oh, this is going to be a great episode. I feel it. I feel it already. I'm going to start off with some rapid fire questions. Are you Chris ? Are you sitting down? I am. Okay. That's not one of the questions I just gotta, you gotta be ready for this. Okay. First of all, best pass me restaurant.

Speaker 2:

Uh, let's go with Jake Lee .

Speaker 1:

Good choice road trip with any celebrity. Okay .

Speaker 2:

Oh, Hmm . Let's say Tom Hanks.

Speaker 1:

It's the best show that is currently streamed.

Speaker 2:

Uh, you know, I'm really looking forward to the next installation. Whenever it might come of stranger things.

Speaker 1:

All right. This one, this one's a little bit of a think piece. Are you ready? What will be obsolete in 20 years from now?

Speaker 2:

Oh man. I'm tempted to say just about everything. Um, but I think laptops,

Speaker 1:

Simon, you work in open technology solutions. What does that mean? And what do you do?

Speaker 2:

You know , it's really just , uh, any of the technologies , um, that are developed open source . Uh, we, we pull those together in whatever works best for our clients to help them achieve their objectives. I'm the service area owner of dev ops. And so as that, I held us , uh, help clients get things delivered quickly, get it, get their technology, to , uh , get their technology to market quickly. That's really about build something, test something and deliver something, right? That's what we help clients do . The way that companies react when they're losing money or losing customers , um, and everything Springs into action to solve the problem, right? The right people get together quickly to pull whatever solution needs to be put in place. And as we think about that, that's really not that different from the goal of dev ops, right? We want to get things done and we're going to get it done quickly, but we also want to do it safely. So we got to get the right people involved to go through the right checks and processes to make sure we don't break stuff or open huge security gaps. And so I just observed that when you're having some crisis as a company that involves technology, you typically pull together a war room to try to solve that problem. And in that war room is everybody needs to be involved in making a change to your production environment. Something that typically can take weeks and months or even quarters to do, right. And yet it starts to get done quickly and with all due diligence , uh, available to solve the problem. So from that I just said, Hey, you know what? There are patterns that we can observe there that you can take and apply to processes that aren't having happening in the middle of an emergency and make it something that you do as a daily discipline about doing , uh , practicing the right rigor with the right people involved to get changes deployed quickly.

Speaker 3:

Well, one aspect of doing business is sometimes the frustration of making a decision one day then having to wait several weeks or even months , uh , for that decision to , uh , take action. So what is it about the war room that is able to solve problems quickly?

Speaker 2:

It's the fact that all of the right people are involved, right? So there's no decision that needs to be made that , uh, needs somebody who's not involved in that process. Right? So the war room has to have everybody who can make the fixed test to fix and deploy the stakes quickly and then validated the fix was made properly. And that's really the power of the war room and moving quickly.

Speaker 3:

So it sounds like the right checks and balances have to exist. What about executive stakeholder buy-in? Is that something that is considered as part of the war room?

Speaker 2:

Yeah, absolutely. And the war room is so clearly focused because the problem is so severe, right? The systems are down. Revenue is being lost. Customers are fleeing the platform if you have the executive attention. Right? So if a similar level of executive presence can be , um, aligned around value added instead of just loss prevention, then there's the opportunity to get , uh, to, to leverage that focus into quick delivery and quick innovation as well.

Speaker 3:

So when the war room is convened Simon, are you looking to swing for the fences and make big changes? Or is it, is the approach a little more discerning ?

Speaker 2:

Yeah. Great question. No, you're not , you're not trying to make big changes, right? Why? Because big changes break stuff in very complex ways. When you pull a Warren together, you're really just trying to make very small changes. Uh , and there's a bunch of reasons for that, right? A small change has a lower likelihood of breaking things. A smaller change is easier to test and a smaller change is easier to deploy. Does it make it faster as well? Yeah, definitely. It's quicker to deploy and is quicker to validate that has done what you want it to do.

Speaker 3:

If a company or organization were to begin planning to convene a war room , uh , to be proactive, what are some of the things they need to be doing as part of that plan?

Speaker 2:

It's a great question. They've got to think about , um, who needs to be involved ideally and moving a change from an idea all the way to running code that your customers are using. This will typically involve the developers to actually build it , uh , the testers to be involved in testing it or automating the testing of that, your deployment engineers who actually deploy it into production, but then also security and compliance. Uh, those people need to be involved to make sure that we're not sidestepping any necessary , um, uh, standards or best practices that have to be in place or creating any security holes. So getting all those people involved, as well as an executive stakeholder to affirm the direction, according to the business objectives is critical.

Speaker 3:

Once that plan is, is put into place, what are some of the next steps to guide that path?

Speaker 2:

Do you want to ask that question before you put it into place? How are we going to know if it's working? Right. Um, so typically what you want to do is collect data that's produced by the change itself to let you know if the change is having the desired effect. For example, if you're putting in place a process that will divert users away from a newly deployed and broken path through the system towards a well known and well understood and tested path in the system, you want to be logging out the fact that users are going down the old path instead of the new buggy path, collect that data and validate that the system is behaving as expected. Okay .

Speaker 3:

When I think about these crisis response procedures , uh, what role does security and governance play in the war room?

Speaker 2:

Everyone is panicking typically in these situations, it's a crisis. So there's a huge temptation to take off all the safeguards , take off all the parameters that we typically put in place, the checks and balances all go out the window. Um, and executives will quickly support this because all they can see is the money bleeding out of their company. It's critical to get a security and governance in place for really two reasons. One make sure that we don't do that right, that we don't go introduce a huge gaping security hole and in the rush to solve an existing problem , um, which creates huge problems down the line. And then two from a governance perspective is make sure that anything that is done that was hasty maybe not wrong, but hasty is fixed later and made compliant, right? So we need to keep track of those deviations from standards, which are sometimes acceptable so that we can fix them in a quiet moment after we've resolved the crisis.

Speaker 3:

That sound indicates that it's time for our listener mailbag. Jeff, what do we have today?

Speaker 1:

Okay, let's take a look. Here we go. Dear blueprint, year after year, after our December global go live event, I walked through the workshop and see endless shelves filled with partially built and unpopular toys, toys that we thought would be in demand that just never caught on what can I do to be less wasteful and more efficient Santa Claus ? How lucky are we

Speaker 3:

That we've actually got a subject matter expert like Simon to maybe answer that question .

Speaker 1:

How cool is it that Santa listens to our show?

Speaker 3:

Well, he's, he's always watching. That's the thing. That's this whole setup,

Speaker 2:

Man. What an honor to get , uh, get involved in this , uh, this little production factory here. Blow the question. Glad I got to answer it. So let's see. What was the situation? It sounded like Santa's had a lot of unpopular, partially completed toys and lots of them. If I think about that, I guess he got started to work on things that he thought would be , uh , popular and then never even got them done. And before he got them done, they had already gone out of style. So thinking about how I would encourage him to address that, you know, I think we'd recommend that he build a small, simple version of the toy and actually get it into the hands of children quickly so that he could get their feedback. Right. If the kid likes a simple version, they'll probably like the next version too , but you know, maybe they'll provide feedback that a , they don't like it. Maybe some part is too sharp and keeps cutting them when they play with it. Right. Or, Hey, this toy is stupid and boring or I don't like the color blue, I guess you'd want to incorporate that feedback into the development of the toy.

Speaker 1:

So it seems like Santa is not iterating.

Speaker 2:

Yeah. It seems like that helps them .

Speaker 3:

What assignment is your advice for him to produce one toy at a time to ensure that each toy produced aligns with the demand?

Speaker 2:

Yeah. You know, I guess if he did that , uh, instead of toys that became unpopular before they were ever used, he could produce a toy and match it up with the Christmas letter that he received and actually asked for it.

Speaker 3:

But in order for him to, to facilitate your recommendation, it sounds like it would require him to get toys into the hands of children throughout the year.

Speaker 2:

Yeah, that's true. You know, I'm not really sure what the deal is with delivering toys all at once. That seems like a really bad idea, right? Just the logistics around that alone seems really, really challenging. You could probably streamline things and be much more efficient if he would deliver toys frequently, maybe to one country at a time. So, you know, what I would recommend is maybe just delivering , uh , a few toys each day and just rotate countries. That's receiving toys. Okay .

Speaker 1:

Simon, thanks so much for being a guest Sante . Thanks so much for being a listener. Paul, thanks so much for co-hosting with me today. If you'd like more information about dev ops transformation, please visit [inaudible] dot com tune in next Thursday .

Speaker 3:

That was the blueprint presented by Cordera to learn more about Cordera and find our latest thought leadership. Visit [email protected] or find us on LinkedIn.