Consumption analysis: 5 patterns in quarter-hourly data

If you're responsible for electricity consumption across multiple company sites, you know the problem: the data sits with the grid operator, arrives as an annual total on the bill, and preparing the EEffG compliance report means copying spreadsheets together. Since Austria's ElWG update in December 2025, quarter-hourly smart meter data is transmitted by default: 96 readings per meter, per day.

This article shows what patterns become visible in that data and what they mean for energy managers, facility managers, and ISO 50001 practitioners.

Why quarter-hourly data? What changes compared to annual totals?

Until recently, most companies had only annual billing data for their sites. One number per meter per year. This was supplemented by Standard Load Profiles (SLP): statistical estimates that model a typical customer type's consumption, not the actual consumption at a specific site.

With the ElWG, quarter-hourly readings are now transmitted to third parties by default. The resolution difference is enormous: instead of one data point per year, you now have 35,040 data points per year per meter.

Austria's EEffG (Energy Efficiency Act) requires "current, measured, and verifiable data" for the mandatory Standardisierter Kurzbericht (§42, Annex 1). Quarter-hourly smart meter data meets exactly this standard: measured, timestamped, from the grid operator.

ISO 50001 additionally requires continuous monitoring of Energy Performance Indicators (EnPIs). Daily quarter-hourly data makes this possible without additional metering hardware.

The key point: The data already exists at your meters. The question is whether you access it.

Pattern 1 — Load profile: When does your site consume how much?

The load profile is the starting point for any consumption analysis: a time series showing electricity consumption across the day. What appears as a single number on an annual bill becomes a detailed picture with quarter-hourly data.

For a company site, the profile reveals: operating hours and shift changes, HVAC cycles, the lunchtime dip, Monday morning startup surges. Patterns you suspected but never quantified.

The weekday vs. weekend comparison is particularly revealing. If an office building shows 60% of weekday consumption on Saturdays when nobody is working, the question is: which systems keep running, and why?

If you manage multiple sites, you can overlay profiles to identify outliers. A site whose consumption pattern deviates significantly from comparable buildings is a concrete starting point for investigation.

Relevance for the EEffG Kurzbericht: The report requires disaggregation of main energy consumers into buildings, production processes, and transport. Load profiles provide the data foundation for this breakdown.

Pattern 2 — Base load: What runs when nobody is working?

The base load is the electricity consumption that persists around the clock: nights, weekends, during company holidays. In quarter-hourly data, it appears as the flat floor that never reaches zero.

For a household, that's the fridge and router, roughly 300 to 400 kWh per year. For a company site, the scale is different: server rooms, ventilation systems, emergency lighting, production equipment on standby, compressed air leaks. In commercial buildings, the base load is often 30 to 50 percent of total consumption.

The night and weekend floor in the data shows whether equipment is being properly shut down between shifts. A particularly revealing test: compare the Christmas week or a factory shutdown with a normal operating week. The difference is pure savings potential. Energy consumed without serving a productive purpose.

Relevance for the EEffG Kurzbericht: The report requires efficiency measures with quantified annual savings potential. Base load analysis provides concrete targets and measurable figures.

Pattern 3 — Peak analysis: Where do consumption spikes occur?

Peak loads are the highest consumption values within a measurement period. Quarter-hourly data reveals which 15-minute interval had the highest demand and whether it was a one-off event or a recurring pattern.

For businesses, peaks matter for several reasons. Those with a demand charge (Leistungspreis) in their electricity contract pay for the highest measured demand in the billing period. Beyond that, peaks determine transformer sizing and required grid connection capacity.

Typical causes: simultaneous startup of multiple machines on Monday morning, HVAC and production running in parallel on hot days, electric vehicle charging during peak office hours.

The practical measure is load shifting: stagger equipment startup, charge EVs overnight, move high-demand processes to off-peak hours. Quarter-hourly data makes the relevant time windows visible.

Note: Not all commercial electricity tariffs in Austria include a demand charge. But even independent of the tariff model, understanding peaks is relevant for capacity planning, grid connections, and evaluating efficiency measures.

Relevance for the EEffG Kurzbericht: Load shifting and peak reduction are classic efficiency measures with quantifiable savings potential.

Pattern 4 — Shift operations and production correlation

For manufacturing or logistics sites with shift operations, quarter-hourly data reveals the energy signature of each shift. Two-shift operations look fundamentally different from three-shift operations. The data makes this measurable.

The key question: does consumption drop proportionally between shifts, or does a constant base persist? That constant portion is optimization potential: lighting, ventilation, and machines that aren't being scaled down.

An example pattern: the night shift consumes 80% of the day shift's energy despite significantly lower production. Causes: lighting and HVAC running at full capacity, machines not being reduced.

When production data is available alongside electricity consumption data, the energy intensity metric kWh per production unit becomes calculable. A core ISO 50001 EnPI.

energiedaten.at provides the electricity consumption data. Correlation with production data happens in the customer's energy management system or CAFM. We deliver one data building block, not the analysis tool.

Pattern 5 — Trends and before/after comparisons

Individual days show patterns. But only over weeks and months do trends become visible: seasonal shifts, the effect of new equipment, operational changes like introducing shift work or commissioning a new production line.

Particularly valuable for energy managers is the before/after comparison. New LED lighting, building insulation, new HVAC system: is the measure delivering the expected savings? With daily quarter-hourly data, changes become visible in weeks, not on the next annual bill.

The EEffG Kurzbericht explicitly requires a four-year trend comparison of energy performance indicators (Annex 1 to §42). Starting automated data collection early builds this baseline from day one.

Practical approach: export data monthly, build trend lines, present them to the COO or CTO with concrete kWh reduction figures. Not estimates, but measured values.

From pattern to action: How do you get the data?

The analytical patterns are clear. The practical challenge for energy managers: meters spread across multiple sites, different grid operators, different portals. Manual data collection doesn't scale.

Options for accessing Austrian smart meter data:

• Grid operator web portals: Manual XLSX download, limited to roughly 50 meters per portal. Feasible for a handful of meters, but with 40 sites across 10 grid operators, it becomes a full-time project.
• Build your own EDA integration: Specialized communication software, server infrastructure, 3 to 6 months of development. The full data journey from meter to system illustrates what's involved.
• Managed data services: Handle grid operator communication, consent management, and data normalization, delivering clean data via API, webhook, or CSV.

energiedaten.at specializes in exactly this problem: enter metering point numbers, the meter holders approve data sharing on their grid operator's portal, and consumption data flows daily. From all Austrian grid operators, normalized into one unified format. For energy managers with multiple sites, the Growing plan (up to 150 meters) is a typical starting point. For more than 150 meters, the Business plan offers unlimited capacity with your own communication infrastructure.

If you don't want to start from zero: the historical data import (available from Starter, annual contracts) lets you request past consumption data retroactively from the grid operator, for a complete baseline from day one.

Conclusion

Quarter-hourly data makes five patterns visible that annual totals cannot reveal: load profiles, base load, peak loads, shift patterns, and trends. Each pattern maps directly to the requirements energy managers must meet: EEffG Kurzbericht, ISO 50001 EnPIs, efficiency analysis, capacity planning.

The data exists. 97% of Austrian electricity connections have a smart meter, and quarter-hourly data is the standard since the ElWG. The question isn't whether the data is there, but how it gets from all your sites into your systems reliably.

Try it with one site: Starter plan, 14 days free, 10 meters included.