ThingWorx is different from most startups. Startups typically begin with an idea hatched from a study of market conditions. ThingWorx, on the other hand, builds on the success of interesting earlier ventures.
In early 2009, ThingWorx’s three founders began discussing how we could bring our experiences in creating innovative manufacturing software to bear on the broad set of businesses where people and “things” (systems and devices) are integral to their product or service delivery. These are markets like manufacturing, utilities, transportation, shipping, energy, and some emerging opportunities like the smart grid, smart buildings, and telemedicine.
To grasp the full potential of our what we are trying to do, you need to know where we’re coming from. Our story begins with the first companies we created, and our aim is take those experiences and successes and apply them to the broader challenges and opportunities inherent in the Internet of Things.
John Richardson was the founder and CEO of CIMNET, which specialized in Manufacturing Execution Systems (MES) until Invensys acquired CIMNET in 2007. Manufacturing execution systems occupy a niche between ERP and plant floor control systems, handling such tasks as production data capture, efficiency and downtime reporting, and production workflow. Perhaps because of their relationship to ERP systems, MES proved to be exceptionally monolithic, complex, and difficult to customize, costing hundreds of thousands of dollars to install and a million dollars to implement, with ROI unknown for at least a few years down the road. Customers would run out of money and enthusiasm long before finishing an MES rollout. Manufacturing execution systems were more often than not a failure for the companies that implemented them.
One lesson we took away from CIMNET and from MES in general was the need to break monolithic systems into pieces and mash up only pieces customers needed in as fast, painless, and inexpensive a manner as possible. The only way to drive adoption within companies is if supporters can quickly point to success stories and if there’s so little risk that it literally pays to innovate. Neither of these things can be said about most enterprise applications, which is why they’re on the way out. Some of the challenges we’re trying to tackle with the ThingWorx platform are these: how can you innovate more quickly, easily, and incrementally than in the past? How can you design a storage model that isn’t rigid but that can provide answers to the questions businesses need to ask? (even questions that were not anticipated at the time of system design). What kind of application development do you need for that?
ThingWorx’ other two founders are Russell Fadel and Rick Bullotta, the former CEO and CTO respectively of Lighthammer. Lighthammer pioneered the field of Manufacturing Integration and Intelligence (MII) before SAP acquired the company in 2005. As the name MII implies, Lighthammer’s original focus was on aggregating and visualizing data from disparate manufacturing systems, with the goal of placing real-time and historical performance data in the hands of whoever could use it and contribute insights for improving operations. This unified visibility approach of bringing more eyeballs to bear on processes became a hallmark of Lighthammer’s first product, Illuminator.
But Lighthammer soon realized there was a significant amount of untapped value in two additional areas. One was integrating the passage of data between systems instead of only people. Lighthammer’s second product, Xacute, offered customers the chance to compose business logic that acted upon those streams of data. The third product was a real-time quality engine, which helped formalize continuous improvement processes. Together, the three products comprised an enterprise manufacturing intelligence platform no competitor could touch. And Lighthammer customers frequently found themselves buying the solution for one piece or functional area, and then leveraging the platform on a wide variety of applications for operational excellence. This generated tremendous value for Lighthammer’s customers and inspired the platform-based approach that lies at the heart of ThingWorx.
We also discovered uses for this software that went far beyond manufacturing. One company’s Six Sigma Black Belts applied it to measuring sales and IT performance because the platform allowed them to aggregate new data sources and deliver analysis to people cost-effectively and in a meaningful way. Another used it to measure its rate of innovation, drawing upon data from its SAP and Siebel systems. And a third used it to optimize its energy purchasing and production mix on a moment-to-moment basis. None of these things fell under the heading of “traditional” manufacturing; instead, it was a pattern of connecting systems to systems and people to systems that proved to be most valuable to them.
And that’s where it stopped after the acquisition. We look back fondly on Lighthammer as a sort of unfinished symphony. While our customers received tremendous business return, there were two areas that we had planned to improve the platform to deliver even greater value. The first was to enable model-based development that would significantly increases reuse and thus developer productivity. The second was to enable business users to perform ad hoc information discovery and analysis to solve business issues.. This isn’t to say that ThingWorx is just Lighthammer 2.0. It’s about applying what we’ve learned and taking that pattern to the next level, creating an entirely new platform that harnesses new technologies like Web 2.0, the semantic web, and model-driven development. We have also designed it to optionally be deployed in the cloud to increase access and reduce costs. The goal is to drive the cost and risks of innovation down to something approaching zero.
We’re not finished improving that pattern, either. A fundamental shift in computing is underway. An Internet primarily composed of people is giving way to an Internet of Things where every device and every system is connected. Because of this, applications will have to change from a hierarchical, pull-driven model dominated by people to an event-driven, peer-to-peer approach with more flexible collection and storage schemas than anything enterprise software offers today. And that’s exactly what we’ve set out to do at ThingWorx. Our application development platform is tailor-made for this new paradigm, in which systems, devices and people are fluidly connected and their interactions are exposed in a searchable and semantic model. This creates an environment where everyone can add value. Business users can use search to find information to solve pressing problems and identify opportunities, and application developers and IT staff are freed up to work on their vision and unique innovations. That’s where we’ve come from, and that’s where we’re headed.