Half-hourly consumption data, refreshed roughly every 30 minutes

Half-hourly electricity, gas and water readings come from the settlement data flows attached to each meter and land in Purely Insights roughly every 30 minutes once settlement has run. Half-hourly electricity meters in profile classes P5 to P8 are read natively at that cadence. Missing reads are picked up by anomaly detection rather than passing unnoticed, so a gap in the data shows on screen as a gap instead of quietly distorting a monthly total.

The customer side of onboarding is short. Sign the supply contract, hand over a portfolio list or upload the latest bills, accept the data-access letters of authority and pick who receives alerts. Contract sign to live dashboard runs 7 to 14 working days, and the platform consolidates every site under one login regardless of who supplies it, so a portfolio split across several suppliers is read in one place rather than one portal at a time.

Not every meter is half-hourly, and the platform does not pretend otherwise. Non-half-hourly meters in profile classes P1 to P4 are covered through AMR and supplier billing data, with AMR refreshing daily. Sub-meters report at whatever cadence the device runs. The compatibility list is HH (P5 to P8), NHH (P1 to P4), AMR, smart, sub-meter and manual upload as a fallback, which means a mixed estate does not need a metering programme before it can be measured.

Where circuit-level detail matters, Purely supplies, installs and owns wireless and clip-on sub-metering at circuit, sub-circuit and asset level. The hardware sits on Purely's balance sheet and the data sits with the customer, so there is no lease and no capex, and the sensors are returned at no notice cost if the customer leaves. A food and drink manufacturer, single site, 12 GWh, used sub-metering on three production lines to isolate a refrigeration setpoint drift, fixed it in one shift and brought baseload down 11 percent year-on-year.

The same dataset rolls up and drills down. Views run from site to building, meter and circuit, and aggregate hourly, daily, weekly, monthly and annually, with drill-down to a single half-hour when a spike needs explaining. Historical data goes back up to 6 years where supplier data flows allow, which is enough to hold a tender, a change of premises and a plant replacement in the same chart. Year-on-year and month-on-month comparison sit alongside the raw curve.

Raw half-hourly totals are not comparable across a cold February and a mild one, so consumption is weather-adjusted using heating degree days and cooling degree days for like-for-like comparison. That matters most at board level, where a rise caused by weather and a rise caused by a schedule fault call for very different responses. Weather adjustment applies to the year-on-year and month-on-month views as well as site-against-site benchmarking, which carries quartile and outlier views for larger estates.

Anomaly detection, alerting, bill validation and carbon reporting all depend on the same half-hourly series. Weekend usage, overnight spikes, missing reads and holiday baselines can only be detected if consumption is visible at half-hourly resolution, because a monthly total hides all four. Configurable thresholds per meter and sub-meter in kWh, kW, £ or kgCO2e fire by email, SMS or webhook, as a daily digest or in real time, with escalation routing. Peak-demand and DUoS red-band exposure is flagged before the charge reaches the bill.

Baseload is the clearest example. Industry typical baseload sits at 15 to 25 percent of total kWh for an office or school, and anything above 35 percent usually points to a BMS schedule or always-on plant fault. Those are industry ranges, not Purely measurements, but half-hourly data is what lets you place a given site against them. A multi-academy trust with 22 schools found 6 sites with weekend baseload above 8 kW, reset the BMS across all 6 and projected a £41k saving for the academic year.

The same series is what leaves the platform. Exports run to CSV, Excel, PDF and board-pack PDF, with a read-only JSON API on request for finance, BI, ESG and building management integration. A cold storage operator with 4 sites and 18 GWh of electricity fed the Insights API into the Flex desk model, moved tranche cadence from monthly to weekly and took a £74k advantage over the prior strategy. Compliance exports cover SECR, ESOS Phase 4, ISO 50001 evidence packs and CDP.