Case study · Industry
Turning six data sources into one cookieless, GDPR-safe analytics platform
Persona-based dashboards and a secure web portal, helping an NHS staff discounts platform grow both engagement and advertiser revenue.
Python
Django
Chart.js
6
Data sources unified: logs, web stats, geo & more
5
Dashboard tabs, tuned to stay fast at scale
Cookieless
GDPR-safe analytics, by design
The client
Our client’s business is a web application where NHS staff can register for discounted deals from big-name brands. To increase click-throughs and advertising revenue, they wanted to analyse visitor behaviour, so they could design better services for advertisers and make the site more attractive to users.
Data for analysis came from application logs, web stats, geographic location, on-site behaviour, newsletter subscriptions and mail campaigns, drawing on a mix of cloud sources and flat files. The client also wanted to offer advertisers limited access to their own data, as an additional revenue stream.
The challenges
- •Multiple personas: a secure registration and authentication process for staff members with full access, and advertisers limited to only the data relating to services they’d purchased.
- •Privacy: with GDPR in mind, visitor behaviour stats for the main site had to be cookieless.
- •ETL design: data ranged from Google Sheets accessible only via SharePoint to cloud sources reachable by API, with batch jobs needing to run on different schedules into an Azure SQL Server database.
- •Geographic analysis: mapping visitors by NHS Trust, with traffic-light colour coding based on visitor numbers.
What we delivered
- ✓An Azure Data Factory integration layer (linked services, pipelines, dataflows and scheduled triggers) automating extract, transform and load into the reporting database.
- ✓An Azure SQL Server reporting datamart, modelled around the business’s subject areas.
- ✓A secure Django/Python web portal, handling persona-based logins for staff and advertisers alike.
- ✓Training and knowledge transfer to the client’s own team.
The outcome
- ✓Data visualisations split across Deals, Newsletters, Activities, Page Types and Audience tabs, tuned at the SQL level so every tab stayed responsive.
- ✓An agile delivery approach meant regular demos and fast client feedback as features were added.
- ✓Development stayed on schedule and in budget, and the site launched successfully as part of the client’s enhanced offering to advertisers.
The founder’s take
“GDPR ruled out cookies from day one, so this had to be genuinely useful analytics without them. That meant more care in the build, but the client never had to choose between growth and doing right by their users.”
Thomas Madden
Founder & CEO, IO Analytics