This series of three articles dedicated to IS governance aims to highlight the importance of technology and data governance at the highest corporate level, operating in a model that is both agile and adaptive. These articles are written through iQo's collaboration with Isabelle SIPMATransition Manager.
Agile practices have become essential
Methods widely used to facilitate digital transformation
The use of agile methods has grown in companies in recent years fordigital transformation , with the main characteristics:
- Progressive development in successive stages (starting, for example, with the development of a website);
- A functional perimeter covering interactions with third parties (customers, external partners, etc.), which are difficult to grasp in terms of needs, and which raise the challenges of digitalization;
- The superimposition of new applications on the existing technical environment (the "legacy", which is fraught with constraints and expensive to adapt), rather than the adaptation of existing applications;
- The use of new technologies (cloud, etc.) and architectures (real-time, APIs, etc.).
The transformation challenges to which these methods must respond are as follows:
- Time-to-market requirements are increasing: technology is becoming a competitive differentiator, and delivering functionalities quickly is essential;
- Difficulty in defining detailed functionalities from the outset: for applications intended for a large number of users, often external to the company, it is not possible to draw up a detailed specification for a finished product that will satisfy everyone, right from the start;
- The need to regularly introduce new functionalities: Gathering customer feedback and adapting functionalities are key;
- The need to constantly ensure a balance between added value and cost: as new perimeters have to be automated, it is essential to contain the cost in order to avoid an uncontrolled explosion in IT expenditure;
Broader needs raise coordination issues
Each agile team can initially enjoy a good degree of autonomy, with limited interaction with other transformation initiatives. The governance rules built into these methods can be applied locally without too much difficulty in small teams of 5 to 9 people, with a light link to more global IS governance.
But today, the scope of these methods has widened, and we speak of " product mode" . "product mode or "or "agility at scale to manage transformation programs involving up to 150 people each. Methods such as SAFe, but also Scrum@scale, Spotify and LeSS, offer a model for managing the transformation of an entire enterprise. What fundamental changes does this imply for a company applying traditional IS governance practices?
Deploying agile methods "at scale" requires a change in governance
A mature methodological repository to guide the transition from agility to scale
The governance required to implement agility at scale is widely described in various methods such as SAFe, which also provide for a progressive approach to implementation, which can be carried out in stages (first a "train", then a program, then a set of programs, and finally a portfolio of initiatives for the entire company). Experience shows that this is a lengthy process (3 to 5 years minimum), and that few companies deploy these approaches across their entire business.
The model described is based on :
- A set of roles and committees has been set up to coordinate several agile teams, replicating the way an agile team operates on a more aggregated level. For example, the Product Manager coordinates a group of Product Owners, and the Release Planning meeting uses the same operating principles as a Sprint Planning meeting for all the sprints making up a Release. Backlog management is also broken down into different levels (e.g. User Story / Feature / Epic), which allow us to model the way in which functionalities are grouped together to form value chains that enable products or services to be delivered;
- Cross-functional teams providing support to agile teams, and acting as guardians of cross-functional coherence. These are essentially the teams ofarchitects, systems teams, and those responsible for quality and methodological support, as well as all the communities of practice;
- A decision-making process on orientations and priorities, as in any governance system: the particularity of agility at scale is that it requires both participation (all players contribute to the decision-making process, to the extent that they are asked to vote at the end of the process), and synchronization (decisions are taken at fixed intervals, usually quarterly, at the same time for all teams and all scopes).
An essential point for successfully implementing agility governance at scale on a subset of the company is to describe the functional scopes to be implemented and distribute them among the teams, taking care to :
- Minimize the coordination workload between teams, and thus the links and dependencies between functionalities;
- Ensure that each team can autonomously deliver tangible elements of value (to understand the conditions for team effectiveness, discover Team Topologies approaches) within the framework of a sprint.
The need to hybridize methods in large, traditional organizations
With the exception of startups and technology-intensive companies, few companies today have deployed agile methods across their entire IS. On the other hand, most companies need to practice agility on increasingly large and strategic perimeters (to find out more about the link between agility at scale and performance, we invite you to read Accelerate).
They all need to develop governance models that enable them to articulate transformations and management mobilizing a variety of technologies, but also different methods and rhythms. This requires :
- Link together transformations carried out using different approaches (agile, V-cycle, etc.) thanks to dynamic management of a portfolio of initiatives at the level of the company as a whole;
- Coordinate functional development needs with IS operational maintenance requirements (incident correction, IS security, service level adequacy, technical debt management, etc.).
Bringing different methods together in a renewed IS governance system
For a company that uses agile at scale alongside other methods (the most common case, as a transition phase or final stable state), a common minimum system must be put in place to guarantee overall consistency. This must not prevent agile methods from operating effectively.
New IS governance based on three fundamental pillars and a Build/Run unification principle
The 3 pillars of governance, which must be managed at the level of the company as a whole and remain common to all methodological practices, are :
- Transformation design , which consists in defining and sharing a common vision of future targets: this element is a powerful glue, enabling coordination through team cooperation around common objectives that make sense. Numerous collective intelligence tools facilitate the co-construction of the target vision: Business Model Canvas, Mom Test, Jobs-To-be-done, Impact Map, Event Storming, Domain Storytelling, etc.
- Steering strategic alignment by allocating resources by value/product chain, not just by project: this means reinforcing value-based steering, by setting ambitious targets that are few in number (using OKR-type approaches), and by regularly measuring the cost/value ratio of transformation initiatives ( "Lean Portfolio Management" method ), in a broad approach to value (not just financial);
- Overall management of the company's capacity to act, which implies specific, long-term management of all internal and external resources: this management must be set up and organized on a company-wide scale to ensure overall resilience.
Finally, the "Build/Run" continuity principle: as the last element of our "New Look" IS governance, it is necessary not to reduce governance to the management of a portfolio of business initiatives, and to consider the IS also in its "Run" aspect, with the requirement to ensure quality service on a daily basis, and to guarantee the security and maintainability of IT tools.
When it comes to "Run today, too, we need to be agile. As the scope of IS use expands to include customers and other third parties outside the company, the service level requirements are those of these new users (customers, for example, use the IS in the evenings and at weekends). These new requirements are imposed on project teams at the development stage, and call for good control of production processes to avoid degrading operation, and a high level of responsiveness in the event of an incident.
The separation between development and production is no longer relevant, and "Run" requirements must be placed on the same level as business functionalities in priority management processes. New governance approaches must therefore integrate this "Run" dimension not separately, but rather as an integral part of all decision-making processes.
Governance leading to a review of the respective roles of the business and IT departments
As we have already mentioned, IS governance in a context of agility at scale relies heavily on the business lines, which take on key roles: the role of "Product Owner" in teams, and the role of "Epic Owner" or "Product Manager" transversally.
But this is not enough. The transition from a mode of prioritization based on project budgets with deadlines to a prioritization of strategic initiatives that are implemented operationally with the collective intelligence of teams represents a governance challenge. Indeed, management must delegate a broad scope of responsibility to the above-mentioned roles, without relinquishing control of the more global impact on their business perimeter. This means reconciling a strategic vision that focuses on the company's major challenges, with an operational vision that includes a detailed understanding of end-users' expectations. This also presupposes a strong involvement in the technological component associated with business changes, and in the management of transformations. This movement is underway, but there is still a long way to go before business units are even more involved in managing the transformations brought about by technology.
For a long time, the Information Systems Department was considered to be solely responsible for the entire Information System, while at the same time suffering from a lack of agility (so-called "shadow IT"). In reality, the IS can only be managed and developed with the participation of all the company's departments, and the IT Department cannot assume ultimate responsibility for it alone. Nevertheless, its role is decisive in guaranteeing overall coherence and efficiency in the management of the IS as a whole. We therefore need to redefine its role in a renewed IS governance system.
Further information
Towards agile IS governance (Episode 3 / 3)
The final installment in our series of three articles on IS governance. These articles are written through iQo's collaboration with Isabelle SIPMA,
Towards agile IS governance (Episode 1 / 3)
This series of three articles on IS governance aims to underline the importance of technology and data governance at its best.
Data governance: definition and key principles
Data governance is a major lever for improving performance and accelerating business transformation. Indeed, today, data is an essential
