Performance intelligence
See how every site really performs.
Totals say what happened. Performance intelligence explains why: fair comparisons, weather and occupancy taken into account, and savings you can rely on.
Method notes- Peer positionTop quartile44.8 kWh per guest night
- Against expected5.7% betterAfter weather and occupancy
- Verified savings380 MWh/yr1,225 MWh/yr measured, kept separate
- Hotspots3 of 14 sitesAbove expected after adjustment
01 Peer benchmarking
Know where each site genuinely stands.
A benchmark is only useful when it is fair. Each site is compared on a sector measure, such as energy per guest night, against comparable assets.
Each dot is one comparable site in the comparison set. Further left means less energy per guest night.
How it works
- 01Pick the measure that fits the sector: kWh per guest night, per m², per bed-day or per tonne handled.
- 02Compare with similar external sites, or across your own portfolio. The two stay separate.
- 03See where the site sits, and how far it is from the next quarter of the group.
What it tells you
- Which sites lead and which lag on a like-for-like basis
- How large the realistic improvement range is
- Where an internal comparison and an external one disagree
How it works and what it tells you
How it works
- 01Pick the measure that fits the sector: kWh per guest night, per m², per bed-day or per tonne handled.
- 02Compare with similar external sites, or across your own portfolio. The two stay separate.
- 03See where the site sits, and how far it is from the next quarter of the group.
What it tells you
- Which sites lead and which lag on a like-for-like basis
- How large the realistic improvement range is
- Where an internal comparison and an external one disagree
Shows where a site genuinely stands, and how far it is from the next quartile.
Benchmarks only hold between comparable assets: similar type, boundaries, services and operating conditions. Comparison sets use relevant reference data, and their make-up is shown with the result.
02 Adjusted performance
Separate conditions from performance.
A mild summer or a quiet quarter can look like progress. An adjusted baseline (what the site would have used in today’s weather and occupancy) gives credit only to real improvement.
STEP 1Metered consumption this year, with last year’s monthly values for reference. On its own, this mixes weather and occupancy with performance.
How it works
- 01Learn how past usage moved with drivers such as degree days (how hot or cold it was), occupancy, hours or production.
- 02Apply that to this period’s conditions to get the adjusted baseline.
- 03Compare actual usage with it. The difference is performance; the rest is conditions.
What it tells you
- Raw change versus last year, as metered
- How much of that change came from weather and activity
- How much came from how the site was operated
How it works and what it tells you
How it works
- 01Learn how past usage moved with drivers such as degree days (how hot or cold it was), occupancy, hours or production.
- 02Apply that to this period’s conditions to get the adjusted baseline.
- 03Compare actual usage with it. The difference is performance; the rest is conditions.
What it tells you
- Raw change versus last year, as metered
- How much of that change came from weather and activity
- How much came from how the site was operated
Credit real improvement, and stop reporting a mild summer as progress.
Adjusted baselines use a regression model (a statistical fit of past usage against the drivers), where enough good data exists. Actual and adjusted figures are always shown side by side. Not every reduction is an efficiency gain.
03 Savings tracking
Track savings by strength of evidence.
A board should know how much it can rely on a saving. Each initiative shows its estimate, its metered result and, where agreed, its verified saving.
READUp to three bars per initiative, one per evidence level. Status sits on the right of each row.
How it works
- 01Record the initiative with its engineering estimate before implementation.
- 02Measure the result against the adjusted baseline once it is in place.
- 03Mark it verified once it passes an agreed measurement and verification (M&V) process or independent review.
What it tells you
- Which savings are expected, which are measured and which are verified
- Where results fall short of estimates
- Which initiatives are worth repeating across the portfolio
How it works and what it tells you
How it works
- 01Record the initiative with its engineering estimate before implementation.
- 02Measure the result against the adjusted baseline once it is in place.
- 03Mark it verified once it passes an agreed measurement and verification (M&V) process or independent review.
What it tells you
- Which savings are expected, which are measured and which are verified
- Where results fall short of estimates
- Which initiatives are worth repeating across the portfolio
Know which savings you can stand behind in a board report or contract.
Estimated: an engineering calculation made before implementation. Measured: metered results compared with the adjusted baseline. Verified: confirmed through an agreed M&V process or independent review. The three are reported separately and never added into one total.
04 Portfolio intelligence
Find where to act first.
In most portfolios, a few sites explain most of the gap. SustainMatrix ranks every site against its expected level and shows the likely cause.
READSites ranked from furthest above their expected level. ▲ above expected, ▼ better than expected. Hotspots carry their main driver.
How it works
- 01Bring every site onto the same indicators and adjustment rules.
- 02Rank sites by how far they sit from their expected level.
- 03Cut the view by property, location, brand or business unit, then drill into the site.
What it tells you
- The underperforming sites and operational hotspots
- The likely driver of each gap, such as cooling, out-of-hours load or kitchen extract
- Where capital and management attention will move the portfolio most
How it works and what it tells you
How it works
- 01Bring every site onto the same indicators and adjustment rules.
- 02Rank sites by how far they sit from their expected level.
- 03Cut the view by property, location, brand or business unit, then drill into the site.
What it tells you
- The underperforming sites and operational hotspots
- The likely driver of each gap, such as cooling, out-of-hours load or kitchen extract
- Where capital and management attention will move the portfolio most
Focus capital and attention on the few sites that move the portfolio.
Rankings depend on data completeness and on the adjustment models agreed for each site type. Sites with incomplete data are flagged rather than ranked.
Method notes
Transparent by design.
These figures shape budgets and public statements. These principles sit behind every number.
Benchmarks compare like with like: asset type, boundaries, services and operating conditions are stated with the result.
Baselines are adjusted for drivers such as degree days and occupancy where data allows. Actual and adjusted figures are shown together.
Estimated, measured and verified savings are labelled and reported separately, never blended into one figure.
Gaps are flagged, not filled silently. A site with incomplete data is marked before it is ranked or reported.
AI assists with analysis. People review the models, assumptions and results before they are used.
Every figure links back to its source document, conversion factor and the person who approved it.
Questions, answered plainly.
What is weather-normalised performance?
It compares this period’s consumption with what the site would have used under the same weather and activity in the baseline period. The difference shows how operations changed, separately from the conditions.
How is peer benchmarking different from portfolio comparison?
Peer benchmarking places a site against comparable external reference assets. Portfolio comparison ranks your own sites against each other. SustainMatrix keeps the two separate because they answer different questions.
What is the difference between measured and verified savings?
Measured savings compare metered results with the adjusted baseline. Verified savings have also passed an agreed measurement and verification process or an independent review.
What data do I need to start?
Monthly consumption by site, plus the drivers that matter for your sector, such as occupancy or floor area. More history and better drivers make adjustment more reliable.
Which reporting standards does this support?
SustainMatrix supports data and reporting aligned to the GHG Protocol, ISO 14064, GRI, ISSB, CDP and SBTi. See standards and frameworks.