Harper, formerly known as HarperDB, has rebranded to reflect its evolution from a database provider to a full-stack web application platform. In an exclusive interview, CEO and co-founder Stephen Goldberg discusses how Harper’s innovative approach fuses database, caching, and messaging into a single process, delivering ultra-low-latency performance that outpaces traditional backend solutions like MongoDB, Redis, and Kafka.
Goldberg explains that Harper’s architecture eliminates data movement bottlenecks, enabling p95 response times as low as 1 millisecond. This performance edge is already driving significant ROI for customers, with one major e-commerce brand projecting a $100M revenue increase due to Harper’s speed.
As online commerce continues to surge, Harper is setting new performance benchmarks, optimizing everything from search rankings to conversion rates. Looking ahead, the company is expanding its focus beyond e-commerce to real-time media and AI-powered personalization at the edge. With its distributed, peer-to-peer mesh architecture, Harper is redefining web application delivery, blurring the lines between cloud and edge computing.
Read on to learn more about Harper’s game-changing technology and what’s next for the company.
Please tell us more about yourself?
My name is Stephen Goldberg, and I am the co-founder and CEO of Harper. I’ve worked in tech since I was a teenager in the mid-90s and have spent my career working in enterprise and infrastructure.
What is Harper and what unique solutions do you provide?
Harper provides a web application platform that combines node.js application development, caching, database, and messaging into a single process. This unique architecture allows developers to write enterprise-grade node.js applications with p95 response latencies in the one-millisecond range. This is possible by having direct in-memory access to database, caching, and messaging directly within their application code and in-process.
Harper has dropped the “DB” from its name, signaling a shift from a database provider to a full-stack application delivery platform. What prompted this rebrand, and what does it mean for your existing and future customers?
From the onset, our vision was larger than database, but we knew from the get-go that database was the first building block to solve. Ultimately, we have always been building a platform to solve certain architectural challenges in massively scaled web applications. We have spent the last eight years building and refining our technology and we find that the problems we solve for customers are more focused on web application delivery and less on traditional database functionality. As a result, our customers already refer to us as Harper and we thought it made sense to rebrand to reflect how the market already sees us and how we have evolved.
How does Harper differentiate itself from traditional backend solutions like MongoDB, Redis, and Kafka? What makes Harper’s unified platform a game-changer for enterprises?
It’s not about being a better operational database than MongoDB, or a better cache than Redis, or a better messaging layer than Kafka—it’s about being a better architecture for highly scaled web applications. Moving data between these layers drives tremendous latency due to serialization and deserialization, handshakes, network latency, and application complexity. By fusing these technologies, Harper can deliver dramatically better performance at a lower cost and with less complexity than existing architectures using multiple systems.
Speed is a critical factor in digital commerce success. Harper claims to process backend operations in just 0.2-1 milliseconds; how does that translate into real-world business impact for brands?
Latency impacting revenue is not a new problem. From the dawn of digital commerce, companies have been fighting an arms race to have the most performant digital commerce platforms. Lower latency impacts search, SEO, conversion rates, customer satisfaction, cart abandonment, product grid placement, and much more.
You’ve cited statistics on how slow load times impact conversion rates and revenue. Can you share an example of a customer who has seen a measurable ROI after switching to Harper?
A major e-commerce customer projects an additional $100M in revenue due to Harper’s speed.
1) Faster performance improves Lighthouse scores, leading to better Google rankings and more traffic.
2) Harper consistently delivers LCP times of 0.2 seconds or less—10x faster than many competitors.
3) Google considers sub-2-second LCP “good,” but Harper is an order of magnitude faster.
Harper has effectively collapsed multiple software layers into a single, low-latency system. How does this architecture enable such high performance, and what are the technical breakthroughs behind it?
Harper, from the beginning, was built on a web application framework by leveraging node.js and distributed computing. Other products were built as standalone vertically scaled applications. As a result, Harper is a web application platform with database, caching, and messaging capabilities embedded.
A lot of vendors have combined these capabilities at some level as a feature set: caching, messaging, etc. Bolting together a bunch of disparate systems does not lead to the performance results you get with Harper.
The problem is caused by the movement of data from system to system. Combining products into a single offering does not solve this problem alone. Engineering a product from the ground up, that allows direct access to the storage layer, caching layer, and messaging within your node.js code is complex and requires an engineering team of Harper’s caliber to solve.
Your platform secures, manages, and extracts value from data at the edge. How does Harper’s edge computing approach compare to traditional cloud-based architectures, and what advantages does it offer?
Harper blurs the line between edge and cloud by seamlessly synchronizing data across locations. This is made possible by the fact that Harper is built on a truly distributed, peer-to-peer mesh architecture as well as being completely agnostic to the infrastructure it runs on. One of the key advantages is reduced latency—placing servers near customers for optimal performance.
Harper presents a single service layer across edge, cloud, and on-premises giving enterprises a reduction in complexity with freedom to deploy where they want.
With online shopping now outpacing in-store shopping, how do you see Harper shaping the future of e-commerce and digital experiences over the next five years?
Harper is setting the new performance standard for e-commerce, aligning with Google’s speed-driven ranking system. Faster performing websites get more clicks, better SEO rankings, and higher revenue—making speed a necessity. At scale, retailers will need Harper or a similar system to remain competitive in the next five years. Speed isn’t just an advantage—it’s a requirement for e-commerce survival.
What’s next for Harper? Are there any upcoming features, partnerships, or industry verticals you’re particularly excited about?
Harper is disrupting multiple industries, with e-commerce as a natural first fit due to its direct ROI from speed. Additionally, real-time video streaming, and other media, is a key focus, with upcoming blob storage to accelerate innovation. Finally, as AI continues to drive innovation Harper’s ability to provide AI powered product content, SEO optimization, and personalization on the edge, in a cost effective and performant manner, will be a major driver in the next year.
Anything you want to say to developers and the people using Harper today?
We appreciate you! The community around Harper has been our favorite part of the journey. We are deeply grateful to the people who have helped make our product better and helped us get to where we are today. We encourage dev teams to join our Slack community.
