CIOs are understandably frustrated. Charged with maintaining IT operational excellence, they’re also under enormous pressure to deliver more value to the business through digital products and services via software delivery. Isn’t that why they invested in Agile, DevOps and then Scaled Agile Framework (SAFe)? To accelerate value delivery? These initiatives were meant to speed up development, deployment and address the issues of scaling. So why is Time to Value (TtV) still too long, unpredictable and unmeasurable?
If you’re a CIO (or any IT executive at a large organization, agency or institution), you may not know why SAFe (and other methodologies) isn’t working, because you can’t see what’s going on; you have no holistic view into the fundamental mechanics of your software delivery value stream.
And if you can’t see how you’re planning, building and delivering software products — you can’t see how work (value) flows from customer request to operation and back through the feedback loop. So how do you find and fix the problem to improve Time to Value?
SAFe — a guide to success, not a magical methodology
At its core, SAFe is about unifying individual Agile development teams who work concurrently on different/multiple features for one product/large project through the Agile Release Train (ART). Urgent/priority/core features are triaged and progressed by individual teams with the aim of delivering an integrated product as quickly as possible to shorten TtV.
Full SAFe represents the most comprehensive configuration, supporting building large, integrated solutions that typically require hundreds of people or more to develop and maintain
No end-to-end visibility into SAFe
Best-of-breed tools house crucial artifacts (business-related, value-adding assets attached to product development) such as requirements, features, tests, defects and trouble tickets, but do not automatically flow these artifacts between them. That means there’s no easy way to get a clear end-to-end view of how the work flows across the value stream — no quick or simple way to identify waste and bottlenecks.
Many tools are involved in the four main phases that make up the activities and processes of software delivery
Consequently, specialists are forced into informal and error-prone manual handoffs through emails, chats, spreadsheets, tool switching, duplicate data entry, etc. Not only is this manual overhead all non-value adding work that is additional waste, but it provides a fragmented view of the value stream.
Key information is locked within individual tools that do not automatically share updates and changes in real-time. As a result, end-to-end value stream activity isn’t linked between dependent stages. There is no one source of truth or clear narrative about a product’s development as it moves through key stages. And even if you extract information from each tool and piece it all together, who’s to say it’s complete, consistent and detailed enough to present the whole accurate picture as it stands in real-time?
Trying to get a real-time insight into the end-to-end flow of work is near impossible. CIOs simply cannot know whether their investments into SAFe, Agile, DevOps etc. are paying off. How do they know if features are being delivered faster to boost revenues?
SAFe in action
By looking at how SAFe works without an integrated value stream, you can begin to understand how it falls down when you have no end-to-end visibility. Assume you’re a CIO who’s invested in SAFe. In theory, you now have a framework and process in place to manage your software delivery value stream and connect it to the business.
Your team identifies and prioritizes a business initiative, then creates a requirement, analyzes it and breaks it down into stories. Your Agile development teams are now unified through ART — with each team working on an important feature — with the aim of using Agile practices (such as Scrum) to deliver it ASAP. Once the code is committed, they test features as part of an integrated product, then releases them into the customer feedback loop.
On paper, the process sounds pretty airtight — yet a product is taking longer than expected, and a customer wants an update. Its future business is a stake. How do you check what’s going on? There’s no SAFe interface or dashboard. So, you’ll need to revert to old school means of communication.
You consult the product owner, who doesn’t quite know what the latest status is, or exactly why it’s running late. He consults the Scrum owners of each individual team. A lot of emails and meetings take place. A lot of time is lost — precious time where the customer isn’t receiving value from their product.
After scrambling around and chasing up everyone involved in building the product — the PMO/project manager, the business analysts, the development teams, testers — a vague picture is built following a lot of buck passing, a lot of guesswork, a lot of, “err…I don’t know!”
The truth is, while SAFe has provided framework for a value stream, you can’t actually see the people, “material” (i.e., the information) and the processes building the product. You can’t easily spot where the bottleneck is or obtain an accurate account of the flow of work. And other products are getting waylaid, with business needs not being met, and a lot of revenue being squandered.
How do you get the visibility, traceability and control over the process to fix and improve it? You need to turn on the lights to see the bottleneck. You need to dig into all the processes and activities that underpin the value stream. You need Value Stream Management.
Value Stream Management — a SAFe scan
Value Stream Management (VSM) enables you to “scan” your value stream and bring all those little details to the surface. VSM does this by forcing you to connect all the stages you’ve mapped out. With no one tool or system to manage a product suite, VSM is the only way to see “SAFe in action.”
Value Stream Management makes it possible to visualize, measure and optimize software delivery
From the business request through to the planning, building and delivery stages to the customer and back to the business, it asks you to think about how work (value) flows from one stage to the next and to identify waste. This is where the all-important knowledge sharing element of software delivery is elevated.
As noted, a disconnected toolchain provides a static obscure picture of what’s going on. There is no end-to-end single view of what’s going on, what’s happened and why the product is late or incomplete. Value stream integration connects all these teams and tools and provides that necessary end-to-end visibility and traceability by visualizing the value stream.
In many ways, SAFe is a rough, generic blueprint of the software delivery value stream and its alignment to business strategies and initiatives. Meanwhile, by connecting the tools that flow of work, VSM exposes the activities and processes that underpin how value flows. This enables you to measure, optimize and build on the foundations provided by SAFe. It picks up where SAFe leaves off and shows you what’s not there.
VSM forces you to face and adapt to the reality of your how your value stream is constructed and helps you to bring to life and modernize the blueprint you sketched out with SAFe. You can now funnel time back into value-adding work and continuously improve the value stream to increase throughput of more lucrative work.
Not only that, but it eliminates most of the manual overhead, improving the speed and accuracy of the work created. For organizations with a typical development team of 1500+, VSM can save up to $10 million per year productivity overhead. And that’s a pretty good way to justify to your CEO the million-dollar investments made in the latest methodologies, tools and specialists.