The machine at the center of Connected Energy's story is not a gleaming battery or a row of servers. It is an irrigation pump sitting beside a field, doing one brutally simple job: moving water. Run thousands of those pumps at once and a utility sees a problem. Move their schedules by a few hours and the same pumps become a resource - a block of electricity demand that can be shaped, measured and offered back to the grid.
Connected Energy built a business in that sliver between the farmer's need for water and the utility's need for capacity. Its CNRG platform forecasts when equipment will run, groups devices into dispatchable fleets, sends control signals, measures the result and reports whether the promised megawatts actually appeared. The farmer gets a phone-friendly portal for checking weather and starting a pump without driving to the well. The utility gets something closer to a small power plant, except the product is electricity not consumed at the wrong moment.
01The internet was supposed to travel through the outlet
The company began in Pittsburgh in 2004 as BPL Global. The initials meant broadband over power lines, a period idea with excellent cocktail-party physics: use the electrical network already connected to nearly every building as the last mile for internet service. In 2005, BPL Global joined a pilot in suburban Monroeville with DQE Communications and Amperion. At the time, power-line broadband was pitched as a rival to cable and DSL, while radio operators argued that the technology created interference.
The broadband promise never became BPL Global's durable center of gravity. But the experiment left behind a more valuable observation. If utilities could communicate over their networks, they could do more than sell internet access. They could monitor transformers, spot outages and control equipment all the way to a customer's premises. The communications pipe became a means rather than the product.
Capital arrived quickly: a $25 million round in 2006, a $26 million Series C in 2007 and another $23 million in 2008. At least $74 million was disclosed across those three financings. The money funded international expansion, software development and a run of acquisitions. One of those deals would eventually lend the whole company its name.
02A 15-person acquisition ate the parent's identity
In 2008, BPL Global agreed to buy Connected Energy Corp., a Rochester company founded by engineer Christopher Campbell in 1997. The target had begun around outsourced compressed air for factories, then developed web software to monitor generators and industrial equipment. By the time of the sale, its system was monitoring or controlling more than 450 megawatts of demand and generation assets for customers including Duke Energy, Southern California Edison and Lawrence Berkeley National Laboratory.
The logic was clean. BPL Global had capital, utility relationships and international reach. Connected Energy had a working control center for distributed equipment. Terms were not disclosed. Campbell had considered several offers after concluding that raising the large follow-on capital needed in Rochester was no sure thing.
BPL Global later sold its Serveron transformer-monitoring assets to Qualitrol and concentrated on demand response and distributed resources. In 2018, it completed the identity swap: BPL Global became Connected Energy. CEO Mark Rupnik said the old name no longer reflected an industry moving toward connected solar, storage, thermostats, vehicles and controllable loads. The product had swallowed the parent, at least on the sign outside.
03What the software actually does
CNRG is an edge distributed-energy-resource management system, or DERMS. Its buyers are electric utilities. Its daily users include distribution engineers, analysts, program managers, field crews and support staff. The assets under control can include smart thermostats, air conditioners, water heaters, EV chargers, batteries, solar inverters, factory loads and, unusually, agricultural pumps.
The company sells more than a control-room screen. It integrates communications and direct-load-control hardware, supports participant recruitment, manages installations, processes incentives and staffs customer service. Its controllers provide device-level measurement; the irrigation material says they sample electrical signals at 4 kHz and can warn of motor trouble. The platform can be hosted in a secure cloud or installed on utility premises.
That bundle explains the business model. This is enterprise SaaS wrapped around hardware, integration and managed services, sold through tailored utility contracts. Public list prices do not exist. A territory with cellular coverage, modern thermostats and thousands of pumps has different economics from one stitching together an old one-way radio network. The work is closer to operating infrastructure than buying a software seat.
The disclosed 2006-2008 financings totaled at least $74 million. Prices for the 2008 purchase of the original Connected Energy, the later Serveron asset sale and CLEAResult's 2026 acquisition were all kept private. Anyone claiming a neat purchase price is guessing.
04The field failed before the code did
A Southern California Edison irrigation pilot shows why this category resists tidy demos. The plan called for controllers on customer pumps, connected through an On-Ramp Wireless network. The project needed suitable electrical loads, radio coverage, willing owners and signed agreements. Delays pushed site recruitment into summer 2015. With a December deadline approaching, the team could not secure the intended customer locations.
What failed first was not dispatch logic. It was the scavenger hunt for a farm that satisfied every operational condition. SCE and BPL Global changed the venue, installing devices on two pumps at the Tulare Energy Education Center, close to an existing access point and utility staff. Tests in December 2015 and January 2016 successfully curtailed both pumps together and separately, cancelled events, accepted an OpenADR command and showed live electrical load. The farmer portal worked on phones and tablets.
The workaround reduced the pilot's realism - an education center is not a commercial grower - but preserved the learning loop. It also exposed the actual hierarchy of risk in grid-edge work: customer agreements and communications coverage can stop a program before an algorithm gets a chance to be clever.
05The moat is mud, measurement and customer patience
Connected Energy competes with in-house utility systems and edge-DERMS platforms from companies such as EnergyHub and Uplight, along with broader grid offerings from Schneider Electric and other industrial vendors. It is not the largest name in that field. Its distinction is the agricultural wedge and the operational machinery around it.
Irrigation is attractive because each participant can contribute a substantial load and pumping can often move within a watering window without harming crops. It is also unforgiving. A utility must respect agronomy, event opt-outs and equipment safety. The communications need to reach rural wells. Measurement must persuade auditors and market operators. Farmers need a reason to tolerate somebody else touching the schedule.
That last point makes the farmer portal more important than it looks. Remote start-stop control, weather overlays and operating history save trips between wells. The participant receives value even when no grid event is running. Connected Energy says program referrals often travel by word of mouth and opt-outs are rare. Retention is not a marketing layer bolted onto the control system; it is one of the control system's inputs.
Start with one flexible load whose owner already understands scheduling. A narrow, useful program beats a generic device catalog.
Marketable capacity requires evidence. Build verification into the controller rather than estimating everything after the event.
Offer incentives, then make the portal useful on ordinary days. Convenience lowers churn when curtailment season arrives.
Map coverage, contracts, installation crews and fallback sites before promising a launch date.
06Where the model breaks
The playbook does not work everywhere. Loads that cannot shift without damaging output, comfort or safety make poor candidates. Sparse assets can erase the economics of installation and support. Weak communications turn a dispatchable fleet into a hopeful suggestion. Programs fail when incentives do not cover the customer's inconvenience, when baselines are too fuzzy for regulators or when a utility lacks the people and systems to act on the data.
Nor is direct control automatically better than price signals or behavioral programs. Hardware earns its keep where the load is large, repeatable and valuable enough to justify a truck roll. For a small appliance with intermittent use, a software integration or tariff may be cheaper. The correct unit of analysis is not the device. It is verified kilowatts delivered after enrollment, installation, communications, opt-outs and support.
07CLEAResult buys the operating knowledge
On August 25, 2026, CLEAResult announced that it had acquired Connected Energy. The buyer already designs and runs efficiency and demand-management programs for utilities across North America. Connected Energy adds event orchestration, direct-load-control hardware and particular strength in commercial, industrial and agricultural markets. The price was not disclosed.
The timing is legible. Data centers are adding enormous blocks of demand while EVs, batteries, solar and smart equipment multiply at the edge. Utilities need capacity sooner than they can always build wires and generation. CLEAResult gets a platform plus a team that has spent years discovering which pumps answer, which radios reach and which participants stay enrolled.
Connected Energy's 21-year path is not a clean pivot story. It is a pileup of pilots, acquisitions, asset sales, name changes and field fixes. That is what makes it useful. The company stopped treating connectivity as the destination and turned it into plumbing. It found a load customers could move, measured the movement and gave those customers something back. The future of the grid, it turns out, may depend less on a grand new machine than on teaching thousands of old ones when to wait.