Portfolio › AI-Enabled Analytics
How to make sense of
3,000 free-text survey responses
The annual survey has gone out and 750 people have responded. The ratings take care of themselves – any tool can chart a 1-to-5 scale. It’s the qualitative data that swallows weeks: 4 questions by 750 respondents is 3,000 free-text comments somebody has to read, code and defend to the executive. Here’s how we solve it with our bespoke sentiment and theme reader. In this synthetic example every response is scored from -1 (negative) through 0 (neutral) to +1 (positive), the themes surface themselves, and the interactive dashboard below is yours to take for a test drive.
28
Days to prep
the data
7
Days to
actionable insights
100%
Insight
coverage
Context
The real surveys behind this demo
This page is conceptual and broadly inspired by QILT. For current data refer to the QILT website.
QILT stands for Quality Indicators for Learning and Teaching. It is a suite of national surveys funded by the Australian Government Department of Education. These surveys collect feedback to help higher education institutions improve, and to provide transparent data for prospective students on the ComparED website.
QILT conducts four surveys across higher education:
Data methodology
The synthetic dataset
Data source
Synthetic dataset
3,000 qualitative (free text responses) by 750 synthetically generated student survey records across 9 schools, 4 questions, inspired by QILT structure over 2 years (2023–2024).
Coverage
3,000 text responses · 750 respondents · 4 questions · 9 schools
Q1 Teaching Quality
Q2 Student Support
Q3 Learning Resources
Q4 Campus & Facilities
Faculty/School · Domestic/International · Undergraduate/Postgraduate · SES, GOS, and GOS-L
Demo
Data dashboard
Loading dashboard...
Architecture
How the pipeline works
End-to-end architecture from survey data to executive dashboard - built inside existing government cloud infrastructure.
Data Preparation
Microsoft Excel
Extract Transform Load
Microsoft Azure
Gold
Analytics-ready
Silver
Cleansed · Validated
Bronze
Raw ingest
Cortex AI
Snowflake
Semantic Model
Access Tiers
Executive Dashboard
Outputs
Executive benefits
Speed
Minutes vs weeks. Hours to build, minutes to refresh.
Consistency
Eliminates manual variability. Same model, every response.
Actionability
Quantified theme insights. Colour-coded by sentiment.
Need qualitative survey data analysed at scale?
Snowflake Cortex AI sentiment and theme extraction, Microsoft Fabric semantic model, Power BI dashboards with faculty-level filtering. Scoped, quoted, delivered.