Why do process modelling initiatives so often bog down, and how does separating process architecture from process modelling fix that?
Process modelling is widely recognised as valuable — yet in practice, many initiatives consume enormous effort without delivering insight. Teams produce detailed models of the wrong processes, or model at a level of technical detail that alienates the business stakeholders whose buy-in is essential. The root cause, this paper argues, is the failure to distinguish between process architecture — a rapid, high-level characterisation of what processes exist, who initiates them, what they produce, and how they connect — and the detailed modelling and design work that follows for selected processes.
The paper presents a structured approach developed and refined by Graham McLeod over decades of enterprise architecture and process consulting practice. Process architectures can typically be captured in a one-to-two hour facilitated workshop, with experienced teams producing four to six per day — a stark contrast to the weeks or months typically consumed by detailed process models. These architectures are held in a repository-backed meta model that integrates processes with business goals, stakeholders, rules, information, applications, and risks, and can be progressively refined into detailed models without rework or loss of fidelity.
Case studies from major financial services organisations — including a life assurer, a private bank, and a merchant bank — demonstrate estimated project time savings of up to 60%, with significantly higher engagement from business participants. The paper makes a compelling case that process architecture, done well, is the essential first step that makes everything downstream faster, cheaper, and more strategically aligned.
Pages: 13
Originally published as a conference paper by Graham McLeod, circa 2008–2009.