← Solutions
Product

DecisionFit — AI Feasibility Platform

DecisionFit reviews a proposed course or qualification and returns a scored feasibility assessment in minutes. Each score is backed by cited evidence and delivered as a PDF your stakeholders can read. It runs on Azure.

Book a demo →All solutions
Delivered in partnership with Techno Union
The problem

Before a university commits to a new course or qualification, someone has to answer “is this actually worth launching?”, weighing student demand, job-market outlook, competitors, cost and accreditation. Today that takes weeks of manual research and often ends in inconsistent, hard-to-defend conclusions.

Our solution

An executive submits a proposal and, in minutes, gets a scored feasibility analysis across every strategic dimension, every claim traceable to its source, as a stakeholder-ready PDF, plus a discovery engine that surfaces new programmes worth exploring.

The environment

It runs in your Azure environment with Entra SSO sign-in, input guardrails and a consistent scoring engine. Your own experts set the dimensions, sources and prompts from an admin portal.

Watch it in action

See the platform run end to end

From proposal to a scored, fully-cited feasibility report, in one continuous walkthrough.

The business problem

Feasibility decisions take weeks

Program feasibility shapes strategy and revenue, but the research behind it is slow, subjective and hard to audit.

  • Manual research across government, industry and institutional sources takes weeks per program
  • No standard methodology, different analysts, different conclusions
  • Market viability, equity and compliance rarely assessed together
  • Citations, rationale and scoring can't be traced or reproduced
Our solution

AI feasibility in minutes

A React portal where an executive submits a proposal, watches a multi-agent AI analysis run live, and downloads a scored, fully-cited PDF, across multiple strategic dimensions.

  • Transparent numeric score (0–100) with written rationale per dimension
  • Every finding links back to a specific document chunk and source
  • Two-layer input guardrails block off-topic or unsafe queries
  • Admins set the dimensions, sources, prompts and scoring from the portal
The multi-agent pipeline

From proposal to cited report

Hierarchical agents, Guardrails → Orchestrator → Dimension Agents → Research Teams → Synthesis, run in parallel on Azure OpenAI. The dimensions are researched at the same time, so an assessment completes in minutes.

STEP 1

Submit & validate

Entra SSO sign-in, then two-layer guardrails check the query before any cost is incurred.

STEP 2

Parallel research

Config loads from SQL; agents research all dimensions concurrently with hybrid RAG search.

STEP 3

Score

Each dimension is scored 0–100, and the AI shows the reasoning behind every score.

STEP 4

Cited report

A final recommendation is written up and delivered as a stakeholder-ready, fully-cited PDF.

How scoring works

How the score is built

The AI reads the evidence, and a fixed engine turns that reading into a number. The same evidence always produces the same score. Every step is explainable and auditable, and your experts can adjust the rules from the admin portal.

LAYER 1

Classify

The synthesis agent labels each sub-criterion, strong, moderate, weak, none or negative, with a cited evidence summary.

LAYER 2

Score

Labels map to values and combine as a weighted mean across each dimension’s sub-criteria.

LAYER 3

Rules

Cap, penalty and red-flag rules apply governance ceilings and flags, caps only ever lower the score.

LAYER 4

Adjust & clamp

Soft adjustments, a 1–100 clamp, and a confidence + viability tag. Dimension scores then roll up with configurable weights.

Labels → scores

strong, clear positive evidence90
moderate, partial, some gaps70
weak, thin or ambiguous50
none, no evidence found40
negative, actively unfavourable20
Nuance: “no issues found” counts as strong; “no data available” is none, not negative, silence is not a penalty.

Worked example, Target Market Viability

Sub-criterionWtLabelScoreContrib.
Employment outlook0.35none4014.0
Enrolment demand0.30none4012.0
Job-market activity0.20none408.0
Competitive intensity0.15strong9013.5
Base score (weighted mean)47.5
Then rules apply: a cap rule can set a ceiling (e.g. if strategy-fit is weak/none → cap at 75), penalties subtract, and red-flags are recorded.

Consistent and traceable

For the same evidence, the engine returns the same score. Any difference comes from how the AI reads the evidence, and the maths stays fixed. The AI explains each score, and every step is traceable. Your subject-matter experts adjust the rules from the admin portal.

Sub-criterion weightsLabel anchorsCap · penalty · red-flag rulesPer-run dimension weightsData-source mappings
Key capabilities

What's built in

Agentic RAG

48,000+ enriched chunks across 17 hybrid search indexes, answers grounded and citable.

Multi-agent pipeline

Up to ~30 agents run in parallel per feasibility, Guardrails, Orchestrator, Dimension & Research agents.

Consistent, configurable scoring

Admin-defined logic that the AI explains for every score.

Two-layer guardrails

Deterministic checks in <10ms plus a semantic model, with a token-budget circuit breaker.

Citations & audit trail

A source behind every claim, with every query, decision and config change logged.

Live progress & PDF

Real-time SignalR dashboard as each dimension completes, then a one-click cited PDF.

The impact

From a live deployment

8Scored dimensions
48k+Enriched knowledge chunks
17Hybrid search indexes
~30AI agents in parallel
100%Claims cited
MinutesPer assessment
WeeksTo deploy in your Azure tenant
AzureMarketplace-bound
See it in action

Inside the platform

A focused React portal where academic executives run a scored analysis, discover new programmes, and configure dimensions and prompts, all in one place.

Feasibility results · Bachelor of Business AdministrationScore 76 / 100
Capability to deliver
65/100
! Moderate

Adequate resources with some investment required.

Equity & access impact
88/100
✓ Good

High prospective student interest and enrolment potential.

Implementation simplicity
78/100
✓ Good

Well-aligned with institutional strengths and resources.

Post-launch scalability
58/100
! Moderate

Existing partnerships with clear room to expand.

Revenue potential
72/100
✓ Good

Positive ROI projected within four years.

Strategic alignment
92/100
✓ Good

Full compliance with accreditation standards.

Course discovery engineAuto recommendations · 6
Ask the engine
What courses should we launch next?
What new courses should we launch in 2027?Which STEM areas have the highest unmet demand?Suggest micro-credentials we could develop quicklyWhat programmes would maximise our existing faculty?
🌐 Market trend96%
Cybersecurity skills shortage is critical, no current programme addresses it

A national critical skills shortage; analyst postings up ~50% in six months. A Graduate Certificate could launch in 8 months using 60% existing curriculum.

Fast to marketSkills shortageHigh demand
↗ Run feasibility: Graduate Certificate in Cybersecurity
🕘 Search history91%
You've analysed Data Science and Nursing separately, but never combined

Cross-referencing past analyses reveals an untapped interdisciplinary opportunity, health data science is projected to add 8,000 new roles by 2030.

InterdisciplinaryFaculty synergyEmerging field
↗ Run feasibility: Bachelor of Health Data Science
Admin · Dimension configuration
Capability to deliver4 sources✓
Equity & access impact4 sources✓
Implementation simplicity4 sources✓
Post-launch scalability4 sources✓
Revenue potential4 sources✓
Target market viability5 sources✓
Implementation simplicity

Evaluates alignment with institutional strengths and academic resources.

DocumentFaculty_Research_Portfolio.pdf
How does this program align with current faculty expertise?
DocumentAcademic_Resources_Inventory.xlsx
Are there sufficient academic resources available?
Edit prompt
Edit Prompt
Update the prompt text and verify with tagged documents
Prompt Text
How does this program align with current faculty expertise?
✦ Get AI Prompt Suggestions
✦ AI-Suggested Prompts
✓What are the key findings and insights relevant to this dimension?
✓Analyse the main opportunities and challenges identified.
✓What data and evidence support the conclusions for this dimension?
✓What recommendations are proposed based on this analysis?
Illustrative recreations of the platform interface.
Under the hood

Built on Azure, end to end

A production-grade stack, every component defined in Bicep, deployed via GitHub Actions with OIDC, scoped to Managed Identity. Secrets are kept out of the code.

Azure OpenAIAzure OpenAIAzure AI SearchAzure AI SearchDocument IntelligenceDocument IntelligenceAzure FunctionsAzure FunctionsDurable FunctionsDurable FunctionsAzure SQLAzure SQLMicrosoft Entra IDMicrosoft Entra IDAzure SignalRAzure SignalRBicep + GitHub ActionsBicep + GitHub Actions
Responsible by design

Accountable, with people in control

Runs entirely within your Azure tenant with role-based access and encryption. Every research run shows its reasoning and confidence, fully logged and reviewable, and a person always has the final say. Aligned with the eight core principles of the Australian AI Ethics Framework.

See it live

Watch it analyse your decision

We'll run the platform on a real proposal and show you the scored, cited result in under a minute, and we can deploy it into your own Azure environment in a matter of weeks.

Get in touch →