The Pyramid of Goodness for Research and Development Organizations
A practical framework for thinking about resource allocation in software companies as a thermodynamic system
--
Introduction
For software engineering startups, particularly ones working on more complex problems, the great majority of capital may be invested in research and development. It is therefore vitally important to have a system of resource allocation that meets several important criteria:
- It provides complete transparency
- The transparency is available continuously in all time horizons
- All stakeholders can access it
- All stakeholders can clearly understand it
- It must be a product of the work itself (i.e. it does not depend on voluntarily reports and such)
Moreover, the system is one of thermodynamic integrity. This is to say, that in all levels of R&D activity, there is no separation between that activity, and the mission of the company.
The Pyramid of Goodness: From Top-Down
Starting from the top, every serious business has a clearly articulated Mission.
The Mission is explained through the Strategic plan.
The Strategic Plan sets the Targets and Budgets for a given period of time.
The Targets and Budgets dictate the Product Roadmap, which in turn influence Engineering Objectives and Maintenance.
These three — Product Roadmap, Engineering Objectives, and Maintenance — contain all the possible activities of R&D.
The Engineering Cycle is the temporal plane in which the work takes place.
Each Engineering Cycle is broken down into Work Orders. The manifest form each Work Order takes is a “Request for Comments” i.e. an engineering spec.
Each Work Order (RFC) is broken down into Pull Requests. These are expressed as checkboxes in the Work Order itself.
Each Pull Request is broken down into individual Commits. In most cases, the commits are expressed as checkboxes in the Pull Request.
The Commits are made of code. Code is the unit of work.
Next, the same, from the bottom-up.
The Pyramid of Goodness: From Bottom-Up
Code makes Commits, Commits make Pull Requests and Pull Requests make Work Orders. Work Orders make Engineering Cycles, and Engineering Cycles make Product Roadmap, Engineering Objectives, and Maintenance. These three make Budgets and Targets. Budgets and Targets make Strategic Plan, and Strategic Plan makes the Mission.
There is no other than that.
Caveats
An astute reader will observe that “engineering work is not only coding”. This is why the checkboxes in Work Orders can have items such as “research so and so technology” or “perform benchmarking tests”.
Roles and Responsibilities
The board and management team are responsible for Mission and Strategic Plan. The CTO works together with management team counterparts and the board to formulate Budgets and Targets (i.e. R&D specific high-level objectives such as “improve energy efficiency by a factor of x”).
The product team provides a Product Roadmap, which the CTO and SVP of Engineering extend into Engineering Objectives and Maintenance priorities.
From there onwards, the subsequent layers of the pyramid are the sole responsibility of the R&D department.
