One of the earliest pieces of sales advice I received was, “It’s better to be lucky than good.” Over time, I learned a more accurate version: timing is everything. Often, success comes down to being in the right place at the right time, with the right solution. Another piece of advice I received was, “Where there’s mystery, there’s margin,” suggesting that if customers don’t fully understand how a solution works, you can justify a higher price. That one I didn’t take to heart. I’ve found far more success offering great products at fair prices. But with the current AI craze, I’m seeing the mystery misnomer take root, convincing people that AI success requires spending millions. In my experience, that’s just not true.
At Progress, we’ve openly embraced the AI revolution —not just internally, but also to support our customers’ initiatives and evolving expectations. Thankfully, our foundation in semantics, knowledge management, artificial intelligence and graph databases positioned us perfectly to adopt and extend the Retrieval-Augmented Generation (RAG) architecture .
We reached out to our existing Progress Semaphore and Progress MarkLogic customers and asked if they’d pilot our RAG solution. Many agreed, and we worked together to refine our design and deliver accurate, relevant results using their private datasets. The collaboration was not only insightful but also validated our technical direction.
Our guiding question: Can a semantically enhanced RAG model deliver meaningful, cost-effective results using an off-the-shelf large language model (LLM)? The answer—confirmed across multiple pilots—was a confident yes.
By integrating customer data into our RAG framework, we increased accuracy from the 70th percentile, even when using traditional vector-based approaches, all the way into the high 90s using the Progress Data Platform . This leap in performance demonstrated not only the strength of semantically enriched architectures but also how impactful it can be to apply RAG thoughtfully with trusted data sources. Even more impressive—aside from the time investment from both teams— the actual costs of these projects remained below the seven-figure mark , proving that enterprise-grade AI doesn’t have to come with an enterprise-sized price tag.
This skepticism isn’t isolated. I recently spoke with a longtime Progress OpenEdge user who couldn’t believe the value packed into their existing setup, especially with the AI enhancements layered in.
At Progress, we’ve always believed in delivering high-quality products at fair prices. That philosophy applies to generative AI, too. If you’re being told your AI success depends on spending millions on software licensing, take a moment to challenge that assumption.
Better yet, reach out. We would be happy to help you test the myth with a pilot of our platform.
Are you ready? Connect with us to get started.
Stephen Reed
Stephen Reed is a Senior Account Executive with Progress. He has over 20 years of technology experience, ranging from artificial intelligence and computer networking to software development and design. Stephen holds a Bachelor of Science in Computer Engineering from Lehigh University and a Master of Science in Information Networking from Carnegie Mellon University.
