Is your organization ready for AI? Take the 10-minute readiness assessment now.
Every franchise location runs on the same playbook. When that playbook is buried across 12,000 documents in a legacy portal that requires keyword searches to navigate, operational consistency becomes harder to maintain than it should be.
That was the reality for a global quick-service restaurant franchisor supporting thousands of locations across multiple regions. Franchisees needed fast, reliable access to product preparation guides, equipment manuals, marketing materials, training resources, and operational procedures. The system they had was making that harder, not easier. The franchisor turned to C4 Technical Services to build something that would actually work.
The franchisee portal held more than 12,000 published documents. To find anything, franchisees typed a keyword and hoped the search returned the right result. When it did, they could solve the problem and keep moving. When it did not, they opened a support ticket. That happened thousands of times each year, pulling support teams away from higher-value work and slowing down answers for locations that needed them immediately.
The consequences of a failed search went beyond a support ticket. Without clear guidance, franchisees could prepare products differently, miss promotional requirements, troubleshoot equipment incorrectly, or fall out of step with brand standards. Inconsistency at the location level creates problems that are difficult to trace and expensive to correct.
AI seemed like the obvious fix. The portal made it complicated.
The system ran on a 2017 Sitefinity CMS with multiple layers of permissions. Franchisees could only view documents approved for their specific regions and product lines. Any AI system would need to enforce those same restrictions at the point of retrieval, not as an afterthought. Returning the wrong document meant exposing information a franchisee was never supposed to see.
The documents themselves created a second problem. More than 20 content types with inconsistent formatting made standard AI retrieval approaches unreliable. The franchisor also had a competing vendor already under evaluation. C4 Technical Services needed to move fast and build it correctly the first time.
C4 Technical Services used its four-phase delivery model to design and build a production-ready AI knowledge system around the franchisor’s content, security requirements, and operational needs.
C4 Technical Services opened the engagement with a technical discovery session involving the franchisor’s CIO, VP of Technology, infrastructure engineering, PMO, and application development teams. The goal was to understand the full environment before making any architectural decisions.
Every document in the portal was mapped by content type, structure, permission level, and quality. That inventory revealed how information was organized, which users could access which content, and where inconsistent formatting was likely to create retrieval problems. Nothing was assumed. The design that followed was built on what the audit found.
Using the inventory findings, C4 Technical Services designed a hub-and-spoke, multi-agent architecture suited to the franchisor’s content structure and permission requirements. A lightweight routing agent receives each franchisee question and directs it to a specialized agent covering operations, products, marketing, or ServiceNow IT support.
This means each question is handled within the correct business context rather than treated as a broad search across the entire portal. Franchisees receive conversational answers with citations linking back to the original document and section, so users can verify the source behind every response.
Critically, permissions were solved before any content was indexed. Authentication and role-based access controls were built into the foundation first. The system only searches content the requesting user is authorized to view. Restricted content is never part of the retrieval pool that generates an answer.
C4 Technical Services built the platform on Microsoft Azure with Microsoft Fabric on the backend. The system handles peak loads of 400 to 500 concurrent users and delivers answers through streaming responses. ServiceNow was integrated directly into the platform. A feedback mechanism was also built in, allowing retrieval quality to improve as franchisees use the system and the organization identifies where answers can be refined.
Audit logging was built into the foundation from day one. Every interaction is traceable, supporting the franchisor’s internal security requirements and its compliance obligations as a Berkshire Hathaway subsidiary.
C4 Technical Services scoped Phase 2 and Phase 3 roadmaps during the engagement, covering Power BI financial reporting, promotional content, brand asset search, and franchisee-initiated actions. Because those future capabilities were accounted for in the original architecture, the production system will not need to be rebuilt as the platform grows. The franchisor received an immediate solution and a foundation designed to expand without starting over.
Franchisees can now ask a question and receive an immediate, cited answer drawn from more than 12,000 approved documents. Each response includes a direct link to the source, giving users faster access to guidance while preserving the ability to verify what the system returned.
Support ticket volume has decreased. Franchisees who previously opened tickets when keyword search failed can now get answers directly from the platform. The support team has more capacity for issues that require direct attention.
The permissions layer means franchisees only ever see information approved for their role, region, and product line. Because access is enforced at retrieval, restricted content is never surfaced in the results used to generate a response. The franchisor did not have to trade easier access for stronger security. Both were built into the system from the start.
C4 Technical Services delivered a production-ready AI knowledge platform that the franchisor could use immediately and scale without rebuilding. The engagement produced results across four areas.
Instant cited answers at scale: Franchisees across thousands of locations can now retrieve citation-backed answers from more than 12,000 documents, with the system handling 400 to 500 concurrent users at peak demand.
Permissions enforced at the retrieval layer: Role-based access controls were solved before any content was indexed. Users only see what they are authorized to see, eliminating data leakage risk without restricting legitimate access.
No throwaway work: The MVP was built on production-grade Azure infrastructure from day one. Phase 2 and Phase 3 capabilities can be added without rebuilding the foundation.
A platform designed to grow: Financial reporting, promotional content, brand asset search, and franchisee-initiated actions are already scoped for future phases, built into an architecture that was designed to carry them from the start.
The franchisor needed a system that could handle complexity without compromising security. C4 Technical Services built one.
If your organization has AI licenses in place but adoption has stalled, C4 Technical Services can build a program designed around your environment.
From improved financial visibility and transparency to maximized ROI, experience the difference FinOps can bring to your cloud investments. Partner with C4 Technical Services and experience expertise that delivers real, measurable results.