The most common billing misunderstanding in Belgian SMEs is mixing up total energy consumption (kWh) with demand intensity (kW). Your monthly kWh can stay stable while your measured peak rises-because in Belgium the capacity tariff is driven by your highest quarter-hour (15-minute) average demand in the month, not by total volume. Fluvius (and other DSOs) record interval data; the single highest 15‑minute average in that period sets your peak and thus a large part of your network and capacity charges. Understanding this explains why "we didn't use more" doesn't stop a higher bill.
Reason 1: Quarter-hour concentration is invisible in monthly totals
Belgian digital meters (e.g. Fluvius) record consumption in 15‑minute intervals. Your billed "monthly peak" is the maximum average power (kW) in any one of those intervals-so a single spike, or several heavy loads starting in the same quarter-hour, can set the whole month's peak. Daily or monthly kWh totals do not show this: you can have unchanged total usage but a new high 15‑minute block (e.g. at opening, after break, or during end-of-day catch-up) that pushes capacity cost up.
Practical fix
Identify your risk windows (first hour, restart after break, end-of-day) and apply a simple stagger rule for non-critical loads so they don't all fall in the same quarter-hour.
Reason 2: Combined "medium" loads create hidden spikes
SMEs often track one obvious large machine but ignore combinations of medium loads. HVAC recovery, compressors, kitchen or process equipment, hot-water systems, and EV charging can overlap and create a larger aggregate than expected. No single load looks problematic alone, but their timing overlap drives tariff impact.
Practical fix
Map simultaneous start events and set a simple sequence offset (for example 10-20 minutes) where operationally possible.
Reason 3: Invoice reading focuses on totals, not capacity components
Many teams only check total amount and kWh. Capacity and distribution charges (driven by that 15‑minute peak) often sit in separate lines or annexes with technical wording (e.g. "capactarief", "piek", "puissance souscrite"), so they get less scrutiny. Result: the invoice "looks normal" while one high interval has already raised your capacity component for the month.
Practical fix
Each month review three things separately: usage (kWh), capacity/network components (and the stated peak if shown), and contract/indexation terms-so you spot peak-driven increases instead of assuming they are "just rate changes".
4. How to get your interval data and verify the peak yourself
Your DSO (e.g. Fluvius in Flanders, Sibelga in Brussels, ORES or Resa elsewhere) holds the raw 15‑minute interval data. You can request access via their online portal or data services. Once you have it, you can identify which interval actually set your monthly peak and at what time-often it will be in the first hour of operation or during a known "all systems on" window. That makes it easier to explain the bill to management and to plan stagger rules with operations.
Some suppliers also show the measured peak (in kW) on the invoice or in an annex. If the number looks wrong-e.g. far above anything you could have drawn-check the billing period and EAN first; if those are correct, request the interval curve from the DSO and consider a formal dispute with the supplier, citing the interval evidence.
5. When to involve operations vs when to dispute the invoice
If the peak is real (confirmed by interval data) but higher than you want, the fix is operational: stagger loads, shift non-critical consumption out of risk windows, and monitor startup sequences. If the peak is wrong-wrong period, wrong EAN, or a clear meter/registration error-then the path is a dispute with the supplier and possibly the DSO, with interval data and invoice references in hand.
Do not mix the two: asking the supplier to "lower the peak" without operational change will not work; and disputing a correct peak without evidence will delay resolution. Get the data first, then choose the right lever.
6. Real-world examples: what a "hidden" peak looks like
Example A: A small factory opens at 07:00. At 07:05 the main compressor starts; at 07:08 the HVAC kicks in to bring the hall to temperature; at 07:10 the first production line is switched on. All three fall in the same 15‑minute interval (e.g. 07:00–07:15). The combined draw in that single interval is 120 kW. The rest of the day the site rarely exceeds 80 kW in any quarter-hour. The monthly peak is 120 kW-and the capacity charge is based on that. Spreading the HVAC start to 07:20 and the line to 07:25 could drop the peak to around 85 kW and cut the capacity component noticeably.
Example B: An office with EV charging. Previously, two company cars were plugged in at 08:00 when staff arrived; the chargers and the building's morning load (lighting, HVAC, IT) sat in the same 15‑minute block. The peak jumped by 22 kW. Moving the EV charge start to 09:30 (or to a timer that begins after 10:00) took those 22 kW out of the peak-forming interval. The next month's invoice showed a lower capacity line with no change in total kWh.
7. Tools and habits that help you stay ahead of the peak
If your DSO or supplier offers a portal with near-real-time or daily consumption and peak data, use it. A quick look at "month-to-date peak" in the second or third week of the month tells you whether you are on track to beat last month or heading for a new high. If you see a spike early, you still have time to remind operations about the stagger rule or to investigate a one-off event (e.g. extra production day, test run) before the month closes.
Keep a simple log: each month note the billed peak (kW), the capacity line (€), and any known event (e.g. "stagger rule introduced", "new compressor", "cold snap"). Over six to twelve months the pattern becomes obvious-which months had avoidable spikes and which stayed under control. That log also doubles as evidence if you ever need to dispute an invoice or explain a cost increase to management.
A repeatable prevention routine
- 1. Flag known high-risk startup windows in operations planning.
- 2. Apply load staggering for flexible systems first.
- 3. Compare expected vs recorded peak before invoice approval.
- 4. Escalate anomalies with interval context and invoice references.
This routine turns capacity tariff management from reactive explanation into preventive control.