Enterprise Management Associates (EMA) recently published their DevOps 2021 research that forecasts nearly 50% growth in Value Stream Management (VSM) vendor penetration during 2020 with a further 20% growth in 2021, and suggest VSM vendors appear to be delivering against market expectations. The slow start speaks to the difficulty in integrating the expansive toolchains that make up the software delivery lifecycle (SDLC).
InfoQ spoke to Jeff Keyes, vice president of product at Plutora, a leading VSM platform provider to understand what VSM adoption involves, how it can have a transformative impact and how it can fail.
InfoQ: How do organisations typically estimate and monitor value delivered?
Jeff Keyes: Most organisations utilise existing project portfolio management (PPM) software to monitor expected value and cost expenditure for major projects and product initiatives. The tracking of the actual value is much less documented, and many organisations lack the maturity to monitor this due to a lack of clarity on the term ‘value’, and what it really means. While most organisations will track improved metrics and cost actuals, not many are quantifying what value really means, therefore making it harder to monitor. This becomes particularly challenging when technology products or services aren’t identified as a value stream. VSM quickly highlights the problems here.
InfoQ: How much confusion do you think there is in the marketplace between the terms value stream mapping and value stream management?
Keyes: There is definitely a lot of confusion in the market between the two terms. With the introduction of agile and lean methodologies into the tech world many years ago, many people have become comfortable with the concept of mapping a value stream and what this means.
Value stream mapping is not process mapping; rather it’s a high-level, bird’s-eye view over the value stream. By thinking of a value stream as a collection of processes we can better understand the goals of value stream mapping. A value stream map enumerates every step of the software value stream, while the mapping exercise identifies processes that need intervention. Value stream management takes this a step further, capturing the flow of work throughout a value stream, identifying how well the value stream is performing and any opportunities to improve it.
InfoQ: How do value stream management platforms aim to join the dots here?
Keyes: VSM platforms bring value stream maps into an ongoing process that aims to continuously improve the value delivered to customers. They do this by operating with an integration model that converges toolchains into a common data model, that then delivers end-to-end visibility and traceability across the value stream. Enterprises no longer need to translate data from one tool to another in a spreadsheet to understand what is going on in the system as a whole. Instead, the deep insights that come from analysing and visualising a wide and diverse data set using the same terms are now accessible, allowing organisations to make data-driven decisions and innovate faster while managing risk. And they manage risk by automating governance, ensuring continuous compliance as work moves through the SDLC.
InfoQ: What are the main challenges you see people trying to overcome by implementing VSM, and where does VSM fail?
Keyes: There are typically four main challenges. Lack of visibility across the portfolio, which means being unable to track all work in progress at a portfolio level; limited visibility throughout the pipeline; poor management and optimisation, which is often found where there are multiple work-streams and different technologies; weak methodologies that mean that even with automated tools, it is not clear to business and IT leaders which metrics should be used as key performance indicators (KPIs).
Of course VSM can fail, like most other practices, if it isn’t integrated well into an organisation’s ways of working. It works best when it’s integrated into both agile and devops to help continually improve software development and delivery. Leadership, development managers, and engineering teams can all benefit from VSM but they will need to dedicate time to incorporate using the platform into their everyday operations to realise its full potential.
InfoQ: What metrics can guide DevOps leaders in moving towards VSM?
Keyes: The metrics that have become industry standard are the DORA metrics. These involve four key metrics: deployment frequency (DF), mean lead time for changes (MLT), mean time to recover (MTTR) and change failure rate (CFR). Organisations should look to adopt these metrics to guide their success, however teams can also begin to look to adopt the more recently developed VSM metrics of lead time, throughput, work efficiency (that is the work item’s active time vs. the total cycle time), cycle time, work in progress, work breakdown (that is the proportion of features vs defects etc.).
Forrester’s Q3 VSM evaluation shows Digital.ai, ServiceNow, Tasktop, and Plutora leading ahead of Targetprocess, IBM, ConnectALL, and CloudBees with Atlassian, GitLab, and Blueprintsys starting to contend in the space as well. InfoQ recently discussed value stream management within the wider context of digital transformation with industry authors Jack Maher and Carmen DeArdo.