About BQE Software
BQE Software is a leading provider of cloud-based business management solutions for architecture, engineering, and professional services firms. Founded in 1995 and headquartered in Torrance, California, BQE has spent over three decades helping project-based businesses run more efficiently.
Their flagship product, BQE CORE, is an all-in-one platform that handles everything from project management and time tracking to billing, accounting, and HR—serving thousands of firms and tens of thousands of users who process billions in annual invoice volume. With a customer-obsessed culture and 24/7/365 support, BQE continuously innovates to help professional services firms succeed.
What challenge did BQE face?
BQE CORE is a comprehensive platform with extensive functionality spanning time & expense tracking, project management, billing, accounting, HR, CRM, payroll, and API integrations. As the platform continues to evolve to serve increasingly sophisticated customers, the need for accessible, immediate support has grown alongside it.
While BQE already had a robust help center, they wanted to provide customers with a more immediate, conversational way to get answers—one that could handle the nuanced, product-specific questions that arise when navigating a powerful ERP system.
BQE needed a solution that could:
- Provide instant, accurate answers to product questions around the clock
- Maintain strict compliance with BQE’s user permission models and feature-specific guidance
- Never assume or generalize when information wasn’t explicitly covered in documentation
- Complement their existing help center with an interactive, conversational experience
How did BQE solve it with CustomGPT.ai? A phased AI strategy
BQE partnered with CustomGPT.ai to build an AI-powered customer experience in 2023, starting with help-desk support and expanding into technical support and then sales & marketing. Their approach demonstrates how organizations can start with a focused use case, prove value, and then expand AI into additional areas of the business.
Phase 1: Product Support (Live). BQE’s AI journey began with the BQE CORE AI Assistant, a context-restricted, highly professional chatbot deployed within their online help center and available in-app via their Resource Center. Powered by CustomGPT.ai’s API, the assistant delivers intelligent responses drawn from verified, tightly scoped documentation. It handles questions across BQE CORE’s full functionality: permissions and security settings; time & expense approvals and workflows; project and phase structures; invoice handling and refunds; CORE API structure and usage; UI filters and navigation. Later in 2023, BQE introduced a further-specialized assistant on their API documentation site focused on their public APIs. Enterprise-grade guardrails ensure the assistant refuses out-of-scope questions, never assumes missing context, and reliably routes users to account managers for package-specific inquiries.
Phase 2: Website Marketing & Sales Chatbot (Live). Building on the support assistants, BQE expanded to the public website with an intelligent chatbot that helps prospective customers navigate BQE’s solutions, answer questions about capabilities, and find resources — a conversational entry point for visitors exploring BQE CORE.
Phase 3: Expanding AI Across the Business (In Progress). With three successful AI deployments in production, BQE is now exploring additional ways to leverage AI within existing workflows — enhancing customer experience, improving internal efficiency, and surfacing insights from customer interactions.
What results did BQE achieve?
BQE’s phased approach to AI has delivered measurable impact: an 86% AI resolution rate, 180,000 support questions answered, and 64% of support tickets handled by AI.
The AI assistants complement BQE’s existing support infrastructure, giving customers an additional way to get answers instantly. At over 100,000 queries annually, the support assistant alone represents a meaningful channel for customer engagement. The documentation team uses the assistant’s analytics to identify patterns in customer questions and continuously improve their knowledge base — creating a virtuous cycle of better self-service outcomes across all channels.
| Metric | Before | After |
|---|---|---|
| Support volume | Human team handling all queries | 180,000+ questions answered by AI |
| AI resolution | Generic or no AI tools | 86% of queries fully resolved by AI |
| Help Center coverage | Static document search only | 64% of interactions handled by AI |
| Availability | Business-hours support model | 24/7 across help center, in-app, API docs, and website |
| Answer grounding | Hallucination risk with generic AI | Source-grounded — verified BQE documentation only |
Why it Worked
- Phased rollout: starting with a focused use case (support), proving value, then expanding to website sales — value delivered at each step.
- Tightly scoped, verified documentation: answers drawn only from approved, in-scope content, with guardrails that refuse out-of-scope questions and never assume missing context.
- Permission-aware guidance: responses respect BQE’s user permission models and route package-specific inquiries to account managers.
- Analytics-driven improvement: the documentation team uses interaction analytics to continuously improve the knowledge base.
Conclusion
BQE’s phased approach — starting with support and expanding to the website — demonstrates how organizations can build AI capabilities incrementally while delivering value at each step. With the support assistant and website chatbot now live, BQE continues to explore how AI can make it easier for customers to get value from the platform.
