Energy baseline and EnPIs in ISO 50001: how to measure energy performance improvement

Published on November 17, 2026

6afcbf297198f_iso-50001-energy-baseline-enpi

An organization can install LED lighting, replace old compressors and run an awareness campaign, and still be unable to answer a simple question: has our energy performance actually improved? If production went up, the weather was milder or a building was sold, comparing this year's energy bill with last year's tells you very little.

ISO 50001 answers this problem with two tools that work together: the energy baseline (EnB) and energy performance indicators (EnPIs). They are the core of what makes ISO 50001 different from most other management system standards — it does not only require a working system, it requires demonstrated improvement in energy performance.

This article explains what ISO 50001:2018 requires in clauses 6.4 and 6.5, the concepts of relevant variables, static factors and normalization, and what auditors check.

Start with the energy review

EnPIs and baselines do not appear out of nowhere. They are built on the energy review (clause 6.3), in which the organization analyses its energy use and consumption based on measurement and other data, and identifies its significant energy uses (SEUs) — the uses that account for substantial energy consumption or offer considerable potential for improvement.

For each SEU, the energy review determines the relevant variables, the current energy performance, and the people whose work affects it. This information is exactly what is needed to choose meaningful indicators and a sound baseline.

Energy performance indicators (EnPIs)

Clause 6.4 requires the organization to determine EnPIs that are appropriate for measuring and monitoring its energy performance and that enable it to demonstrate energy performance improvement. The method for determining and updating EnPIs must be maintained as documented information.

EnPIs can take different forms, for example:

  • measured energy value — total kWh consumed by a site or a piece of equipment;
  • ratio — energy per unit of output, such as kWh per tonne of product or kWh per square metre;
  • statistical model — a regression that relates energy consumption to one or more variables, such as production volume and outdoor temperature;
  • engineering model — a calculation based on the physical characteristics of a system.

A simple ratio is easy to understand but can mislead. If a plant has a large fixed energy load — lighting, heating, compressed air leaks — the kWh per tonne will fall automatically when production increases, even if nothing has been improved. This is why ISO 50001 pays so much attention to relevant variables.

Relevant variables and static factors

ISO 50001 distinguishes between two kinds of factors that affect energy performance:

  • a relevant variable is a quantifiable factor that significantly affects energy performance and changes routinely — production volume, heating and cooling degree days, operating hours, occupancy;
  • a static factor is an identified factor that significantly affects energy performance but does not routinely change — the size of a facility, the design of installed equipment, the number of shifts, the type of product.

The distinction matters because they are handled differently. Relevant variables are used to normalize data. A change in a static factor may require the baseline to be revised.

The energy baseline (EnB)

Clause 6.5 requires the organization to establish an energy baseline using the information from the energy review, taking into account a suitable period of data. The baseline is the quantitative reference against which energy performance is compared — the "before" in a before-and-after comparison.

The choice of period matters. It should be long enough to cover the normal operating cycle, including seasonal effects; twelve months is common, but not mandatory. It should also be representative: a period that includes a long shutdown, a major failure or an exceptional order will distort every comparison that follows.

Where data indicate that relevant variables significantly affect energy performance, the organization must normalize the EnPI values and the corresponding baselines. In simple terms, normalization answers the question: what would we have consumed this year, under this year's conditions, if we had performed as we did in the baseline period? The difference between that figure and actual consumption is the improvement — or the deterioration.

When the baseline must be revised

ISO 50001 requires the baseline to be revised in the following cases:

  • the EnPIs no longer reflect the organization's energy performance;
  • there have been major changes to static factors;
  • according to a predetermined method.

For example, if a factory adds a new production line or closes a building, comparing performance against the old baseline would be meaningless. The organization must retain documented information on the baseline, the relevant variable data and any modifications.

Auditors pay attention here because baseline revisions can also be used, intentionally or not, to make performance look better. A revision should be justified by a real change and applied consistently.

What auditors look for

  • Traceability from the energy review. Do the EnPIs cover the SEUs identified in the energy review, and are the relevant variables those identified for each SEU?
  • Appropriateness of the EnPIs. Do the indicators actually reflect energy performance, or do they move with production volume regardless of what the organization does?
  • A documented method. How were the EnPIs and the baseline determined, and how will they be updated?
  • Normalization. Where relevant variables significantly affect consumption, is normalization applied? Is the statistical model reasonable, and is the data used consistently?
  • Data quality. Clause 6.6 requires a plan for energy data collection. Are meters calibrated or verified as appropriate? Are there gaps or estimated values?
  • Baseline revisions. Were they justified, documented and made according to the defined criteria?
  • Demonstrated improvement. Clause 10.2 requires the organization to demonstrate continual energy performance improvement. For certification, the auditor must be able to verify it — and under ISO 50003, the rules for certification bodies, demonstrated improvement is a condition for certification decisions.

Common mistakes

  • Using only total consumption as an EnPI in an organization where production varies significantly.
  • Choosing a baseline period that is not representative, or changing it every year.
  • Defining EnPIs at site level only, so that improvements and deteriorations in individual SEUs cancel each other out.
  • Claiming improvement based on estimated savings from projects rather than on measured EnPI values.

The ISO 50006 guidance standard describes in detail how to establish EnPIs and baselines, and ISO 50015 covers measurement and verification of energy performance. Both are useful companions for implementers and auditors.

Frequently asked questions

Can an organization have more than one baseline?
Yes. Each EnPI is compared with its corresponding baseline, so an organization with several EnPIs — for different SEUs or sites — will usually have several baselines.

Does ISO 50001 set a minimum improvement target?
No. The organization sets its own objectives and energy targets. What the standard requires is that improvement can be demonstrated against the baseline.

Want to go deeper? Read how ISO 50001 helps companies improve performance. And if you want to prove your knowledge, the ISO 50001 EnMS practitioner and ISO 50001 EnMS auditor certifications are smart options.