Introduction

Description: "Pace-Layered Application Strategy is a methodology for categorizing, selecting, managing and governing applications to support business change, differentiation and innovation, [based on the rate of change of the application]". Via Gartner

Pace-layering differentiates the rate of change into three categories - Systems of Record (slow change), Systems of Differentiation (medium change), and Systems of Innovation (high change).

Goals: This methodology is used to facilitate discussions around:

  • how new and existing initiatives in a portfolio can be best categorized according to the impact and value of it in an organization

  • allows decision makers to assess what kind of investment to make in the different categories

  • the categorization can help to show what kind of governance, project, and product lifecycles are best applied to the initiatives in a portfolio

  • amount of governance required for things at different pace layers

  • what is required to support applications and services in each of these layers; for example, how is innovation supported with easily re-usable data services?

  • the categorization also supports conversations about the appropriate rate of change for systems in each layer, both from a consumer perspective (ie, services that will evolve more or less quickly) and from a support perspective (how do we best architect to manage a level of change appropriate to the layer the system is in).

Context: Once, all portions of a portfolio are classified, we can answer questions around:

  • Is the mix correct (do we have the right amount of innovation vs. record)

  • Is the ongoing investment correct?

  • Is the initiative being governed correctly?

Scope: This is used both at the service portfolio, and the application portfolio level.

Source: Pace layers originated in built architecture with Stewart Brand in 1994, and were adopted by Gartner as an IT methodology.

Scenarios

Pace Layering can be used to guide user expectations.  For example, services could be classified by layer and publicized to users so that they can expect change to happen and an appropriate pace.

Pace Layering can be used to manage applications from an internal-to-IT perspective.  For example, applications can be grouped according to layer and then governance, investment, and development practices can be aligned accordingly.

Method

Roles: 

  • Architect - Lead
  • Senior Leadership - Higher level decision makers
  • SMEs - Contributors and direct decision makers (e.g. Portfolio Owners)

Steps: The steps are designed to produce an end result that should be reviewed and updated iteratively to ensure the recommendations continue to be a valuable product of the process.

Phase I: select/inventory “chunks” (services/technologies/applications) of portfolio that should be classified

  • Led by architect.  Contributors are SMEs and other interested parties

Phase II: classify chunks by the 3 Pace Layers

  • Led by architect.  Contributors are SMEs, and direct decision makers (Portfolio Review Board, for example)

Phase III: workshop for senior leaders to discuss and decide:

  • Is the mix correct (do we have the right amount of innovation vs. record)

  • Is the ongoing investment correct?

  • Is the initiative being governed correctly?

  • what is the appropriate level and means of governance for each layer?

Phase IV: share out/market recommendations that come from phase III

Phase V: hand off to PMO for execution (best practices for governance, investment, and system development)

Communication

  • Diagram of chunks and pace layers

  • Recommendations for changes (maybe a future state of the Diagram of chunks and pace layers)

  • Textual descriptions of recommendations

Examples

(to be completed)

Related Methods

  • Roadmaps

  • Capability Maps

  • Core Diagrams


Architecture Methods > Pace Layers

Contributors

Want to help with this page? Please see the Method Contributor Guide.

Stewards for this page:

  • José Cedeño, Oregon State University
  • Jason Myers, University of Washington

Other contributors:

  • Chris Eagle, University of Michigan

  • Rick Tuthill, University of Massachusetts Amherst

  • Robert Guthrie, Washington University, St. Louis

  • Piet Niederhausen, University of Washington

  • Luke Tracy, University of Michigan

  • J.J. Du Chateau, University of Wisconsin-Madison

  • Jason Myers, University of Washington