Decidr, ICON partner on SAP AI to automate Procure-to-Pay

Australian AI software firm Decidr has expanded into Singapore and announced a partnership with ICON Consulting Group, a SAP services provider in the Asia-Pacific region. The two will work on automating Procure-to-Pay processes for organisations running SAP ERP systems.

Decidr also published findings from a Singapore survey of business decision-makers and outlined an architectural framework it calls the Agentic Graph. It presented the framework this week at the U.S. Capital Access forum.

SAP channel

The partnership will see Decidr's software deployed alongside SAP implementations managed by ICON across multiple countries. ICON works with large organisations and mid-market firms in Singapore, Japan, Australia, Vietnam and India, and has delivered projects for Fortune 500 organisations, the companies said.

Procure-to-Pay will be the initial focus, covering invoice ingestion, vendor validation, purchase order matching and invoice posting. Joint solutions are expected to be generally available in the second half of calendar 2026.

ICON is positioning the work as an extension of existing SAP environments, rather than a rip-and-replace programme. Christophe Derdeyn, Managing Director of ICON Consulting Group, said the target is high-volume back-office tasks.

"The ability to automate high-volume processes like Procure-to-Pay using Decidr's agentic platform represents a significant step forward for our clients. This partnership positions organisations to realise the benefits of AI-enabled operations within their existing SAP environments," Derdeyn said.

The announcement comes as SAP continues to invest in AI features across its product suite and customers weigh how far to adopt automation in finance and procurement. SAP reports more than 425,000 customers across 180 countries.

Singapore findings

Decidr also released research on AI adoption in Singapore. The Decidr Singapore AI Readiness Index Report 2026 surveyed 511 decision-makers across SMEs and enterprises. Nature, a research firm, conducted the study in February 2026.

The core sample covered 401 decision-makers at organisations with 20 to 500 employees and revenue above SGD $200,000. A separate enterprise sample included 110 decision-makers at organisations with 500 or more employees.

The results suggest broad uptake of AI tools, but slower progress on integrated AI operations. Seventy-seven per cent of Singapore businesses said AI significantly enhanced their operations over the past year. Respondents also reported urgency around spending, with 50 per cent of SME decision-makers and 67 per cent of enterprise decision-makers rating AI investment as an urgent priority.

Despite that momentum, only 6 per cent of SMEs said they operate a centralised AI platform across the whole business. The report also found that 70 per cent of AI value still comes from general assistants and copilots, rather than autonomous end-to-end execution.

Expectations of difficulty were similar by company size. Sixty-five per cent of SMEs said they expect AI implementation to be difficult, compared with 67 per cent of enterprise respondents.

Agentic Graph

Decidr's Agentic Graph is positioned as a response to what it sees as a gap between experimentation and execution. It describes many current deployments as fragmented, with tools operating at individual or workflow levels without structural connections across the organisation.

The Agentic Graph is positioned as a coordination layer designed to link organisations, workflows, operational knowledge and decision systems into a single interoperable network. Decidr said organisations adopting its DecidrOS encode processes, decision logic and compliance requirements into a governed schema that connects them within the graph.

The approach reflects a wider debate over where long-term differentiation will sit as foundation models and general-purpose tools become more widely available. Decidr argues that value will accrue to orchestration and governance layers that convert model outputs into actions in business systems, with traceability and auditability.

David Brudenell, Co-CEO of Decidr, said discussions in Singapore have centred on execution and structure, rather than enthusiasm alone.

"Singapore has the ambition, the investment and in many cases now the dedicated teams. What the data tells us is that the countries and companies that win will be the ones that build structurally and execute properly. That's the conversation we're having in Singapore this week, and it's exactly what the ICON partnership is designed to accelerate," Brudenell said.

Decidr was founded in 2018 by Paul Chan. It markets its software as an operating system for goal-oriented strategies, using AI agents integrated into business workflows. It has cited customers including CareerOne, Go1, ELMO, Edible Beauty, Growth Faculty and AIM.

Work with ICON will initially focus on procurement and finance automation in SAP environments. Broader expansion is likely to depend on early deployments and customer demand across the region.