Case studies

CASE STUDY

Patient-centred multi-site digital transformation and AI search solution for NHS Berkshire Healthcare Foundation Trust

The Trust’s digital setup had split into separate sites. This created technical debt, left accessibility gaps, and meant 1 in 3 users could not complete key tasks.

We rebuilt* the digital estate as a single site on Umbraco CMS. Patient referral flows were redesigned to match real user needs. We added RAG* and vector search so users could find answers quickly, even for complex queries.

The result is a secure, compliant site that speeds up patient care and makes it easier for teams to manage content across locations. The approach is now being shared as a model for other NHS trusts.

*(RAG) Retrieval-Augmented Generation. It is an AI framework that improves the quality and accuracy of Large Language Model (LLM) responses by grounding the model in external knowledge sources.

*Growcreate case study

600+

Pages rebuilt around patient journeys

120+   

NHS services standardised for care access

Reading level standard applied across all copy

33%

Increase in task completion

CASE STUDY

From 6% to 64% ChatGPT coverage for Growcreate

When we applied this framework to our own agency, Growcreate—a website and software agency—we wanted AI tools to recommend them to clients.

We audited their content, refreshed key pages, and published targeted authority articles.

The result? Within two weeks, ChatGPT coverage jumped from 6% to 64% of relevant searches. No months of waiting for traditional SEO.

These results are not guaranteed for every engagement. Your outcomes will depend on your categories, sector, your competitors, and your current technical setup. We will always give you an honest timeline based on your unique situation, not just a number meant to make a sale.

*Growcreate case study

64%

ChatGPT coverage after optimisation

14    

Median time to impact

10x 

Increase in brand mentions

#2.1

Average AI rank across top intents

CASE STUDY

Azure OpenAI integration and automated CMS editorial workflows for Mortgage Advice Bureau.

The client needed a way to speed up publishing and keep content secure and compliant across multiple sites.

We added the Growcreate AI Toolkit to Umbraco, connecting Azure OpenAI directly to the CMS. This automated content and metadata, gave editors AI-powered suggestions, and linked up workflows with CRM and cloud tools.

Editorial teams can now publish faster, keep content on-brand, and control costs. The AI tools fit into their existing workflow without adding risk.

*Growcreate case study

50%

Reduction in content publishing time

100% 

Brand and regulatory compliance

Zero

Impact on content approvals

40%

Lower editorial effort

Also trusted by

Mercedes Benz social graphic
CASE STUDY

AI-powered multilingual content localisation solution for Mercedes-Benz

Updating content in multiple languages across global markets was slow and manual. This delayed launches and made it hard to maintain brand consistency.

CASE STUDY

Umbraco website for FE fundinfo, backed by SLA support and maintenance

Technical debt had stalled FE fundinfo's progress, and template and CMS issues undermined content consistency for large editorial teams. Growcreate rebuilt the solution on Umbraco 13, delivering a clean, easy-to-use digital experience. Then they handed the site over to their ISO 27001-certified 24/7 support team for round-the-clock incident response, patching and monitoring, uptime up to 99.95% and critical-issue acknowledgement within 15 minutes. Keeping the site stable, secure and continuously improving.

CASE STUDY

Umbraco platform recovery and modernisation for a global professional services consultancy

Growcreate stabilised and modernised a global professional services and programme management consultancy's legacy Umbraco implementation. Refactoring components incrementally while the live site kept publishing. The result was predictable, low-risk change and faster delivery of enhancements. A strong example of ongoing support and modernisation that avoided the cost and disruption of a full rebuild.