Energy Management Has a Terrible User Experience Problem

We built an entire industry around EPEX prices, EMS platforms, load balancing algorithms, and smart connectors. And somehow forgot to ask the one question that matters: does anyone actually want this?

The gap between what we built and what people want

Talk to anyone in the energy transition and you’ll hear the same pitch. Dynamic contracts. Day-ahead EPEX prices. Setpoints. Charging schedules. APIs between your battery, your EV charger, your heat pump and your inverter, all orchestrated by an “intelligent” EMS that promises to optimize everything.

Now talk to an actual homeowner. They want three things: the car charged in the morning, the house comfortable, and a bill that doesn’t sting. Somewhere down the priority list, if there’s room left, they’d like to feel slightly good about the planet too. That’s the whole list. Four items. Not forty.

What I actually experienced trying to “optimize”

I have solar, a home battery, and an EV. On paper, that’s a perfect setup for dynamic optimization. In practice, I spent a Sunday afternoon configuring charging windows around EPEX day-ahead prices, only to have my battery discharge into the grid at 14:00 because a price signal told it to, right before a cloud rolled in and my afternoon meeting ran on grid power I’d just sold back for four cents. Then the EV charger ignored the whole schedule because it was following its own app, with its own login, with its own idea of “smart.”

Three systems. Three logics. Zero coordination. And I’m the guy who’s supposed to understand this stuff.

Why the current model doesn’t scale

This is not a me-problem. It’s a design problem, and it’s structural.

• Dynamic pricing rewards complexity, not comfort. EPEX prices change hourly. Optimizing against them means someone, you, or software acting on your behalf, has to make dozens of micro-decisions a day.
• Connectors don’t agree with each other. Your battery, your charger and your thermostat come from different vendors, each with its own app, its own assumptions, and no shared goal. Research into smart home interoperability confirms this isn’t a fringe complaint: 73% of smart home device owners rate interoperability as important, precisely because it so rarely works in practice (Security Info Watch).
• Congestion is now part of the equation, and nobody explained that to the user. With regional connection freezes and curtailment becoming standard, most of the province of Utrecht has been under a pause measure on new electricity connections since 1 July 2026 (Energietransitie Utrecht). Your system now also has to behave differently depending on what your local grid can handle that hour. Try explaining “afnamecongestie” to someone who just wants their car charged by 7 AM.

None of this is hype. It’s real complexity, solving a real grid problem. And we handed all of it directly to the end user, dressed up as an “app.”

Built by IT people, for IT people

Here’s the part nobody in this industry likes to admit. These systems weren’t designed around what a household needs. They were designed around what an engineering team could build and what a retailer or platform could monetize.

That shows. A generic-interface study out of KIT scored its energy management dashboard well on usability, 79 out of 100 on the System Usability Scale, but even that positive result came with a footnote: the interface met every requirement except one, system configuration (Xu et al., Energy Informatics). Configuration is exactly the part nobody wants to do, and it’s exactly the part that got left unsolved. A separate usability thesis on smart home systems reached a blunter conclusion after comparing academic research with real deployments: the gap between marketing promises and actual usability is consistent enough across studies to call it a pattern, not an accident (Lavu, TU Wien).

And it’s not only a usability failure. It’s a control failure. Your data, your consumption pattern, your battery behavior, your car’s charging curve, all of it runs through platforms whose business model isn’t “make you comfortable.” It’s “sell you a dynamic contract,” “keep you inside our ecosystem,” or “resell your flexibility to a grid operator.” A 2021 architecture paper on smart home energy management put it plainly: as households become prosumers, the whole point of the system should be that data ownership and access stay with the user, not the platform (SINA architecture, arXiv). In most consumer products today, that’s not how it works. You don’t own the optimization. The retailer does.

Is it really a smart home, or a smart owner keeping it alive

There’s a term for what I was doing that Sunday afternoon: digital housekeeping. Researchers have been studying it since at least 2015, and the findings are consistently uncomfortable for an industry that sells “effortless.” One review describes the concrete components of this labor plainly: setting up devices, mapping the house, fixing recurring glitches, and burning hours on tutorial videos just to learn what a device can do (Digital housekeepers and domestic expertise). That’s not automation. That’s a second job with no salary.

The research goes further. Studies on smart home adoption directly ask whether this labor actually saves time, and the answer is contested at best: some researchers argue digital housekeeping doesn’t reduce domestic work, it just moves it and renames it (Journal of Housing and the Built Environment). A separate ethnographic study makes the same point about the person doing the maintaining: they often enjoy it, treat it as a hobby, and are reluctant to admit it isn’t actually saving anyone time (When Smart Technologies Enter Household Practices).

So the honest answer to “is your home smart” is usually: no, you are. The home just has more things that can break, drift out of sync, or need a firmware update at the worst possible moment. The system didn’t get smarter. The owner got recruited, unpaid, as the systems integrator.

Why switching should never be the price of wanting something new

Ask any homeowner why they haven’t upgraded their EMS, added a second EV, or switched energy suppliers, and you’ll hear some version of: “because then I’d have to redo everything.” That’s not caution. That’s lock-in, and it’s a well-documented one. Analysts covering energy IoT platforms describe the real risk plainly: high switching costs combined with operational inflexibility, so if a vendor raises prices or stalls on features, you’re stuck migrating hardware and data with no easy path out (Vendor Lock-In in Energy IoT Clouds). Industry research on making home energy management work for consumers names the same failure mode from the other side: even where devices are technically interoperable, someone still has to hold control authority, or multiple systems end up issuing conflicting commands to the same charger at the same time (Energy Systems Catapult).

I don’t want to pick a new chargepoint operator every time I want a new feature. I don’t want my battery vendor and my EMS vendor fighting over who gets to decide when my car charges. I want the system built around me, not around whichever vendor got there first.

The real problem isn’t the technology

The algorithms work. The hardware works. Batteries genuinely can shift load, EVs genuinely can charge smarter, EMS platforms genuinely can talk to the grid. The technology is not the bottleneck.

The bottleneck is that we built tools for engineers and handed them to people who never asked to become energy traders in their own living room. That’s not a technology problem. That’s a design problem, and the industry keeps solving it with more settings instead of fewer.

What good looks like: one dial, not ten dashboards

Look at the concept above. Four modes. One dial. Sustainable, affordable, maximum return, or reduce grid congestion. That’s it. No EPEX curve to read. No connector to configure. No setpoint to negotiate with a battery that has its own opinion. No vendor to babysit.

You pick a priority. The system learns your behavior, keeps your data yours, and does the rest: charges the car, runs the heat pump, times the battery, talks to the grid. You’re allowed to not know or care how, and you’re allowed to add a new device or switch a supplier without starting over.

The uncomfortable part

The reason most energy management tools look like a cockpit instead of a thermostat isn’t a technical limitation. It’s that showing complexity feels like showing value, and locking you into one ecosystem is good for the vendor’s retention numbers. More settings looks like more control. It isn’t. It’s just more work handed to the wrong person, in service of the wrong goal.

Most people will never care about netcongestie, EPEX, or load balancing. They will care about a charged car, a comfortable house, a lower bill, and increasingly, a system that’s doing right by the neighborhood without asking them to understand why or hand over control to do it.

Build the dial. Keep the complexity under the hood, keep the data and the goals with the user, not with the retailer.