Industry

Energy & Utilities Software Development

Meter data, billing and field work that agree.

Utility systems deal with two things most business software does not: very high volumes of time-series data, and a workforce that spends its day in the field with intermittent connectivity.

Billing sits on top of that data and has to be defensible. A customer querying a bill is entitled to an explanation, and the explanation has to reconcile with the readings and the tariff that applied.

We build meter data handling and validation, tariff and billing calculation, customer portals with consumption views, field service scheduling and mobile tooling, and the reporting that operations and regulators require.

Common challenges

What tends to go wrong

Meter data arrives in volume and is not clean. Readings are missing, duplicated, out of order or implausible, and a billing run that ingests them uncritically produces bills that cannot be defended.Tariffs are complicated and change: time-of-use bands, standing charges, tiered rates, seasonal variation and regulated price caps, each with an effective date. A bill must be reproducible under the tariff in force at the time.Field work happens where there is no signal, and it cannot wait. An engineer at a property needs the job details and needs to record the outcome regardless of connectivity.

How we address it

What we build instead

Meter data is validated on ingest against plausibility and continuity rules, with anything failing routed to an exception queue rather than into billing. Estimated readings are marked as estimates throughout, so a bill can always say which figures were measured.Tariffs are versioned data with effective dates, and every bill records the tariff version used, making any historical bill reproducible line by line.Field applications are offline-first: the day's jobs are available locally, outcomes and photographs are recorded and queued, and pending items are visible to the engineer rather than silently held.

Outcomes

What typically improves

Billing disputes become answerable with the actual readings and tariff, rather than requiring a manual reconstruction each time.Field engineers complete and record work without depending on a signal, and the office sees outcomes as they happen rather than at the end of a shift.

Frequently asked questions

Can you handle smart meter data volumes?

Yes, though the design differs from ordinary transactional data. High-frequency readings suit time-series storage with aggregation strategies for the queries people actually run, since almost nobody needs half-hourly granularity at a customer level except for a defined recent window.

How do you handle estimated readings?

By marking them as estimates everywhere they appear, and by keeping the actual reading when it arrives alongside the estimate rather than replacing it. A customer asking why a bill changed is entitled to see that an estimate was later corrected, and that requires both figures to exist.

Will field tools work without coverage?

They have to. Jobs are downloaded ahead of time, outcomes recorded locally and synchronised when a connection returns, with clear indication of what is still pending. Anything else results in engineers keeping paper notes, and the data arriving late and incomplete.

Building for energy & utilities?

Tell us how the operation runs today. We will tell you where software would actually help.

Start a project Talk to our team