Report
Document cover

A well-architected framework for public procurement: upgrading institutional capacity in the age of AI

Publisher
Capacity management Quality management Government procurement Public sector innovation Artificial Intelligence (AI)
Resources
Description

Adaptability and agility, accuracy and accessibility should be the features of modern public procurement. A well-architected procurement system optimises institutional design to enable these features. These capabilities are especially urgent for public institutions to acquire while generative Artificial Intelligence (AI) develops at exponential speed into the next general-purpose technology. 

This paper proposes a framework for public procurement, inspired by software architecture, to make procurement more adaptive, responsive and purpose-driven. It identifies four key enablers to implement a new procurement framework. A well-architected procurement system offers the following advantages. 

  • Adaptability – procurement must shift from rigid rule-following to dynamic risk management, ensuring governments can continuously update their capabilities.
  • Agility – instead of one-time purchases, procurement should be designed as an iterative process with feedback loops that refine and improve over time.
  • Accuracy – procurement models should match the characteristics of what is being purchased. Buying cloud computing should not follow the same process as buying office furniture.
  • Accessibility – procurement systems should be transparent, open, and navigable, reducing barriers for start-ups and new entrants while ensuring fair competition.

Four key enablers

  1. Rules – shift from compliance-driven complexity to flexible, outcome-focused regulation.
  2. Culture – promote a problem-solving mindset that embraces risk management, not avoidance.
  3. Talent – elevate procurement to a strategic function, equipping specialists with skills in negotiation, technology, and market analysis.
  4. Data – use analytics and AI to track performance, detect inefficiencies, and improve decision-making.
Publication Details
Access Rights Type:
open
Series:
White Paper #002