VSM the Easy Way: 3 Steps to Get You There

Welcome to "VSM the Easy Way."

We've had a lot of interest around the topic, a lot of interest around VSM and how to approach VSM. How can I make it real for my organization?

So for starters, if you're familiar with VSM, stand up.

And that's a handful. Hang on, stay standing. Stay standing. And if you understand VSM and don't think it's overwhelming, sit down.

That's a handful. And then we've got a bunch of people who I think are like so many of us who don't fully understand and think of it as overwhelming. Now we can sit down. One more poll here.

By a show of hands this time: how many have a VSM initiative going at your company, or had a VSM initiative going at your company?

I'm guessing about maybe 20% of the audience.

So my goal in the next 25 minutes is to demystify VSM: what is it, why, where's the value? And there is tremendous value in VSM. There's real opportunity for competitive advantage. But more importantly, let's get to the how of how we can make VSM real and show a win and move on from there.

So: what is VSM? Why do we want to do it? What's the reality today that is dogging so many companies where VSM could be valuable? And then the majority of what I want to talk to you about is how do we approach it. I've got a three-step method I'm going to present to you, an exercise that you can do, and then what you're going to need from a platform point of view in order to make it real and make it come alive at your company, and what you can do to get started.

VSM, as you may know, is not a new thing. It started a lifetime ago with the Toyota Production System. It started in manufacturing and it has revolutionized most industries in the world, with the notable exception of us in IT and digital. Normally we are the leaders in management innovations, but in the case of VSM, we are late adopters. Toyota and the Japanese auto industry revolutionized the auto world. Every area of manufacturing now operates VSM or associated principles. Famously, Tim Cook, when he came to Apple, implemented VSM principles and first won the favor of Steve Jobs by doing that, to great success.

But it's not just in the physical world of manufacturing. Restaurants like Starbucks, and now so many hospitals and emergency rooms, ERs and so forth with surgeons, all operate according to VSM. We're the laggards.

And that's about to change. I believe that you're here in this room, I'm going to guess you kind of believe that.

The analysts are telling us that this will be the case. The VSM Consortium -- I've got some colleagues from the VSM Consortium here in the room. We had a great event on Monday down the street called Flowtopia. There is so much that is just ready to burst, in a good way, about VSM.

And I'll tell you why. So let's understand what it is.

It brings together the different areas of the SDLC. And in a way, there's nothing new about that. What's new about VSM is it's a very prescriptive approach to how you're going to manage this multi-phase process. And in particular, you're going to manage it by focusing on the flow of value, hence value stream management. And that word flow is so important: making sure that things do flow and don't come to a stop. We don't have wait time.

Speaking of wait time, one of the other prescriptive elements of VSM is an obsession with reducing waste. There's seven or eight forms of waste; wait time is one of them. We want to eliminate all forms of waste, keep things flowing through this multi-phase business process that we have historically thought of as the SDLC. And when we do that, we're going to maximize business value. We're going to deliver more value, better, at higher quality for the business than otherwise. That's what VSM is, and I can't emphasize enough the middle part of utilizing this prescriptive process and fixating on flow and reduction of waste.

If you're just bringing together different parts of the SDLC and doing it the same old way, that's not VSM.

Value streams: this, after all, is value stream management. There are two types of value streams: business value streams and digital value streams.

Business value streams -- I've got some examples right here -- are, as the name suggests, how we deliver value to the end customer, or maybe at the back door to suppliers with supply chain, or even internally with systems that we're using for employees. And this notion of business value streams -- and they're all revolving around systems, typically cloud-based systems -- very much aligns with something I'm going to guess we all believe in, I certainly do, which is the transition from projects to products.

And so we can think of these business value streams, or BVSs, as each of those revolving around a product: a product in this case being your e-commerce system, or your supply chain system, or your appointment-setting system, all starting and ending with the customer. These are examples of business value streams, all revolving around a digital product.

What we're here about today, and what I'm here to talk to you about, is not business value streams per se, but rather: how does the product that powers that business value stream get created and enhanced? That's through a digital value stream.

So a digital value stream is what we might have thought of as the SDLC. And by the way, what SAFe calls the development value stream. I felt that development value stream was too narrow of a term, and like the term digital value stream. And by the way, business value streams is what SAFe calls operational value streams. But again, I think it's more resonant to call them business value streams.

So anyway, digital value streams are going through what we've thought of as the SDLC, with agile going on in the middle. The digital value stream is owned by a CTO or a VP apps or somebody like that. The idea: you've got a feedback loop going with the general manager who owns the business value stream, and life should be good.

Unfortunately, today's reality is not clean. Today's reality is waiting, wondering, and wasting a lot of time and a lot of effort. We've got a bunch of silos going on. We've got various tools in our toolchain going on. And people are pulling out their own spreadsheets and having the classic meetings where everybody's arguing about who's got the right data and so forth. All sorts of negatives from this, including the fact that the business feels like they're not being heard, or they're not getting the status on what's going on. And especially we've got a lot of wait time. You see it at the bottom there: wait, wait, and ultimately we're late to the business.

Everybody lives this reality. How can we get out of this reality?

So again, we want to get to something more smooth like this. And we can think of the digital value stream as going through five canonical phases: plan, build, test, deliver, run. Now, they're not smooth like that. We're not in a waterfall world anymore. There's agile going on, and you sort of see the agile loops there in the middle, in the plan, test, and deliver phases. But something like this, where it is smooth, where we're accommodating the natural recursions of agile and feedback from the business, and where all the different resources that are involved in your DVS, your digital value stream, are part of this and can see what's coming at them from upstream, what's being depended on from them downstream, and we're going to get more, better flow and reduce waste.

So the three steps to get there: let's focus on one digital value stream.

For starters, don't try to do your entire IT operation. Pick one digital value stream and therefore focus on one product. Maybe it's your supply chain system. Maybe it's the checkout subsystem of your e-commerce.

Fixate on that one BVS, that one business value stream, and then look at the digital value stream that feeds it and then profile it.

So what we suggest, and frankly what at Micro Focus we help customers with, is a workshop. Very tight workshop: six, eight hours. That's why it says one day. Profile that digital value stream. And I'm going to show you what the result of that profile looks like in a moment.

You hear a lot about value stream mapping. I'm not against it per se, but what I'm arguing here is in a digital value stream, we sort of know what the standard phases are. We can start there, focus on a single digital value stream, and profile it to understand what's going on, what our tooling is, and so forth. From that, we can move to phase one, which is instrument. And then from there, phase two, accelerate. Phase three, orchestrate.

So let's start with phase one after we've profiled that first DVS, and understand, amongst other things, the tooling that we're using. We're using Jira, we're using ADO, we're using Jenkins, et cetera.

What is being captured there, and what metrics or what data can be extracted so that we can put that into a VSM dashboard and start to operate by looking at flow metrics and waste metrics and other VSM sort of metrics? Then you've got your first big step up into the world of VSM.

So I'm going to show you more about what I mean by that in a moment, but that's the instrument phase. Typically what that then highlights is for certain work areas along your digital value stream, whether it's agile, whether it's what you're doing in functional testing or performance testing, there are deficiencies, and there's too much waste going on, or insufficient coverage, et cetera.

And so what we tend to find then is there's opportunities to accelerate those different phases along the way. And this would be step two: by improving your tooling for those and improving the processes, while still looking at it from a VSM point of view from above with the sort of metrics I just talked about. And then ultimately to orchestrate all of this for better effect.

And after you've done this for your first digital value stream supporting that first chosen product or business value stream, then you look to do it for the next adjacent digital value stream, and from there beyond.

So what I'm introducing here are some simplifying assumptions, some simplifying constraints, so that we don't try to do a full transformation. I wouldn't even use the T word. Start with a single DVS behind a single BVS. Profile it, instrument it, accelerate it, and orchestrate it. Get that early win, get that quick win, and build on from there.

So when you profile it, I've got an eye chart here for you. And obviously it's not meant to be read, but I just want to give you a sense of what we do, and what we recommend you do -- we can help you with it -- in profiling a DVS.

So you sort of profile the overall DVS: who's the owner? What's the business value stream and the product that it's supporting, and who in the business owns that, and who are they competing with, and so forth. And then you drill down to the phases. And you see the phases of the DVS up top: plan, build, test, deliver, run. And profile a dozen things about what's going on at each phase, including: what's your tooling? Where are you using Jira? Where are you using Clarity? Where are you using ADO? What are the issues there? What's the desired state?

This is a very doable exercise in roughly a day's worth of work, with different people along the way. One person's not going to be able to fill this whole thing out. But the head of the PMO probably can be interviewed to fill out the first column, and the head of Dev can be interviewed to fill out the second column on the Dev phase, and so forth.

Having done this, now we're prepped to move ahead to that instrument phase that I showed a moment ago. So that's what you're going to come out of this profile one-day workshop with: it will prep you for doing the instrument step as your next step.

In the instrument step, what we want to be able to do is extract VSM-relevant metrics like flow and things that would contribute to waste from that underlying tooling that's causing the issues identified in the profiling exercise, and feed those metrics into a VSM dashboard.

So that begs the question: VSM dashboard, where would I get one of those, or what would that look like? What do I need to think about in order to be able to do that?

That's going to come from a VSM platform, which would typically have three layers: have an orchestration layer at the bottom, have an acceleration layer in the middle with the acceleration modules I alluded to a moment ago, and then have an analytics or dashboard layer for instrumenting at the top.

Here's our VSM platform. I'm sure there's others out there also. We call our VSM platform ValueEdge. It's an AI-powered cloud-based VSM platform. And in particular, I draw your attention to the top, insights, and you'll see there it's a dashboard that spans your full digital value stream from plan, build, test, deliver, run.

Where the data is going to come from initially: the data typically is going to come from your existing tooling. You see at the bottom you might have Jira, Jenkins, ServiceNow, ADO, Git, GitHub, whatever. That's where your data is going to be coming from, and you bring it in through an integration hub such that it can be displayed up in your VSM dashboard -- which in this case is Insights -- so that you can start to operate via VSM principles and looking for areas where flow is taking too long, or you've got interruptions in flow, waste is too high that you can reduce.

By the way, this is also where AI is very helpful, because it's going to be a lot of data. And what we found is AI and ML is very helpful in crunching that data and identifying areas where waste can be reduced so that you can maximize flow going through.

So that's how you get started with the visibility of those metrics from your existing tooling. And then you can move ahead to step two to identify where you need to accelerate, where maybe your tooling is insufficient or incapable, and you can start to think about the acceleration modules to insert in to improve some of those areas.

So speaking of the dashboard, let's take a look at what the dashboard might look like. Here's an example of the dashboard showing flow metrics for a given digital value stream, including up in the upper right, you're seeing suggestions for where to reduce waste. Those were AI- and ML-generated, by the way, because they're crunching the data of all these metrics that are coming in here and suggesting some things for you to think about where you've got considerable amounts of waste that can be reduced.

And then you see you've got the flow metrics up on the left: flow velocity, flow load, flow distribution. And you're going to start to get some insights. One of our users found they needed to reduce the batch size. When they put smaller batch size into a sprint, they got better flow and ultimately better throughput. That's the kind of insights that you're going to get from a dashboard like this with the sort of AI- and ML-generated insights from it.

So I'm leaving some room for questions here. Hopefully you've got some questions you can start to think about, but I'll sort of summarize here.

Get started with VSM the easy way. Simplify, simplify, simplify. Don't try to do it for your entire operation, for your entire IT operation, or if you've got multiple products -- most companies who are here at DOES have multiple products -- pick one, ideally one that's high impact. Focus on that, and then profile that first DVS, the first digital value stream against that first product. And then get visibility: instrument that, feed the metrics from the existing tooling into your VSM platform or into a VSM dashboard. Now you can start to get the sort of insights that I showed you on the previous slide, and now you're operating via VSM.

Then you want to remove bottlenecks. Bottlenecks are going to become more apparent, and you're going to be able to see them and make the business case for where those bottlenecks should be removed, perhaps by improving your testing capabilities or improving how you're managing agile and your scrums and so forth. We'd love it, of course, if you remove those bottlenecks with one or more of the acceleration modules, but however you go about to do it, your next step is removing bottlenecks.

And then continuously improve. Another principle around VSM that we all learned from Toyota a million years ago is continuous improvement. And so in the orchestration step, you're going to be continuously improving in order to deliver more value to the business faster.