Energy data collection challenges: why it is harder than it looks

One building, one meter, one connection, and the data comes in and keeps coming in as soon as you connect. That is the assumption. Set it up, and it works. The reality of energy data collection across a real estate portfolio comes with challenges that most people do not anticipate. Here is what you encounter.

You do not always know what is there

Before a single data point can be collected, someone needs to know what needs to be measured. That turns out to be the first challenge.

"In many cases, our clients do not know exactly what the local situation looks like," says Kees van Alphen, Managing Director at Hello Energy. "We request photos and only then do we see: it is not one water meter, but sometimes five. There is a main electricity meter, and then a large number of tenant meters that nobody has ever mapped."

The reasons are structural. Energy contracts are often held by tenants, not owners, which means the owner has limited visibility of what is installed. Buildings change over time: meters are added, replaced or moved without documentation. Portfolios are acquired from previous owners without complete handover files.

"We are working on a building in Germany where one tenant used to occupy the whole space and now there are 30 tenants," says Van Alphen. "The infrastructure has been completely changed over time. Nobody has mapped it."

New buildings are simpler, because everything is documented from the start. Older portfolios carry a legacy that only becomes visible when someone actually looks.

Five sources, five kinds of complexity

Once you know what needs to be measured, the next question is how to get the data. The answer depends on what kind of metering infrastructure is in place, and that varies significantly across buildings and countries.

"You have smart meters, dataloggers, BMS systems, invoices and additional sensors," says Benno Schwarz, Head of Growth at Hello Energy. "Each has its own level of accuracy, its own access logic and its own vulnerabilities."

Smart meters are the preferred route: a digital connection, no physical installation, data flowing automatically. But access to those digital connections varies enormously across Europe. "In Denmark, there is one platform where all electricity data for commercial properties is available," says Van Alphen. "We need an identification code and a password, and ten seconds later we have the data. We have spent three years trying to access a comparable platform in another country, eventually gave up, and found a local intermediary instead."

When digital connections are not possible, dataloggers fill the gap. When meters are not smart at all, someone has to go on site, which introduces its own set of delays and dependencies.

In 60% of cases, the process is straightforward. In the other 40%, surprises accumulate. A recent project turned up not one water meter but ten. "For the best possible data quality, we could give every one of those water meters its own datalogger," says Van Alphen. "That is technically possible, but it becomes very expensive. So we go back to the client, because perfect data is rarely the goal."

Getting the data in is step one

A working connection is not the end of the process. It is the beginning of a different one.

"The bulk of our digital connections and dataloggers works without problems once it is set up," says Van Alphen. "But there are all kinds of reasons why it gets interrupted."

Authorisations are the most common cause. "An authorisation is granted temporarily, which means you have to obtain it again every year," says Van Alphen. "That is often a process with complications, because someone does not prioritise it, or because a tenant has decided to consolidate their data with a different metering company. The flow stops, and we have to work out what happened."

Physical changes in a building can break a signal without warning. A meter can be replaced. A supplier system can be updated. By the time anyone notices, days of data may already be missing.

Four energy types, one dataset

If the previous challenges are about access, this one is about what happens after the data arrives. Electricity, gas, heat and water are all measured differently, in different units, using different conversion factors that vary by country, by supplier and sometimes by contract.

"The entire market has moved towards kilowatt hours as the universal unit, so that buildings can be compared and benchmarked," says Lennert, Technical Lead at Hello Energy. "For electricity, the conversion is straightforward. For gas, it is considerably more complex, because the energy intensity of gas depends on its composition, which varies by country and by mix. So a cubic metre of gas in the Netherlands releases a different amount of energy than a cubic metre of gas in France."

The consequences of getting those conversions wrong are not always visible. "Water meters use conversion factors that are often incorrectly configured," says Van Alphen. "The consumption figures look plausible until you manually validate them and realise they are nonsense."

Bringing four different energy types together into one coherent, comparable dataset requires constant attention to these details. A figure that looks correct on screen may have travelled through several conversion steps, any one of which could be wrong.

Which energy data quality do you need?

By this point in the process, a pattern should be clear: energy data collection involves far more variables than the initial assumption suggests. The final layer of complexity is perhaps the most important, and the most overlooked.

"100% data availability is almost always impossible," says Schwarz. "But the question is not whether your data is perfect. The question is whether it meets the requirements for what you need to do with it."

Different purposes demand different standards. Certification requires data that is available at the right moment and meets the threshold for the relevant framework. Analytics benefits from granularity and frequency. Billing requires verified accuracy, with calibrated meters that can withstand scrutiny. Tenant-facing data needs to be correct enough that no one looks incompetent.

Knowing what you need, before you start collecting, changes every decision that follows.

________________________________________

This article is part of a series in which Kees van Alphen and Benno Schwarz share what ten years of European energy data collection has taught them.

‍