Purpose of this article
This article explains how Scaler’s AI data tools support ESG data quality, analysis, and action. It outlines where each tool lives in the platform, what data it uses, and when to use it.
Overview
Scaler’s AI data tools support users across the full ESG data lifecycle — from data input and validation, through analysis and forecasting, to decision support.
These tools leverage asset- and portfolio-level data already stored in Scaler to:
- Improve data accuracy and consistency
- Reduce manual review effort
- Highlight risks, outliers, and opportunities
- Provide actionable insights directly within the platform
All AI features operate fully within Scaler’s secure infrastructure.
Luc van de Boom, CIO, gives a tour of Scaler’s AI features:
- AI Sustainability Advisor: Lumi → min 3:50 & 10:00
- AI Analysis (Graph Insights) → min 6:13
- AI Roadmap Measure Suggestions → min 7:47
- AI Data Quality Report → min 11:10
- AI Forecasts → min 12:44
- Bill Scraping & Verification → min 13:50
AI sustainability advisor (Lumi)
Purpose
To allow users to query portfolio data and Scaler documentation using natural-language prompts, directly within the platform.
Location
- Home
- Analytics Portal → Portfolio → Asset List (Lumi search bar at the top of the table)
What Lumi can do
- Answer questions about portfolio performance, risks, and trends
- Explain metrics, scores, and calculations used in Scaler
- Surface relevant Knowledge Base content and platform guidance
Example prompts
- “Draft a sustainability summary comparing 2023 and 2024 data.”
- “Which assets have the highest transition risk?”
- “How does Scaler calculate data coverage?”
How it works
Lumi references your portfolio dataset and Scaler’s internal documentation to generate context-aware responses.
Responses are stored temporarily and never leave Scaler’s secure AWS environment.
AI analysis (graph insights)
Purpose
To automatically interpret charts and generate contextual insights, explanations, and next steps.
Location
Analytics Portal → Portfolio → Analytics
Select a chart → click AI Analysis (bottom-right of the chart)
What it does
- Analyses the underlying data used in the selected graph
- Produces a short, contextual explanation
- Highlights focus areas or anomalies
- Suggests potential follow-up actions
Supported graphs
- Energy use intensity
- Total energy consumption
- Energy data coverage
- GHG emissions intensity
AI Roadmap measure suggestions
Purpose
To identify and recommend potential efficiency or decarbonisation measures automatically.
Location
Data Collection Portal → Portfolio → Roadmaps → Add measure → AI Measure Suggestion button
Data inputs used
- Energy, fuel, and heating consumption
- Solar production
- Building age and type
- Location and climate zone
- Historical roadmap measures
Outputs
- Suggested measures with estimated EUI or GHG reduction
- Indicative cost and payback estimates
- Alternative options (e.g. insulation, heat pumps, window replacement)
Actions
- Apply suggestions directly to the roadmap
- Create a group of measures for an asset
- Generate a carbon-saving measure
- Request partner advisory support for validation or deeper analysis
AI data quality report
Purpose
To automatically identify anomalies, outliers, and inconsistencies before reporting or audits.
Location
Analytics Portal → Portfolio → Asset List → Data outliers tab
Click AI Analysis at the top of the table
What it does
- Scans datasets at portfolio level
- Flags high-risk assets with severity indicators
- Identifies likely causes (e.g. meter issues, inconsistent readings)
- Suggests corrective actions
Best use case
Run before generating reports to support data integrity and audit readiness.
AI forecasts
Purpose
To estimate missing or incomplete consumption data at meter level.
Location
Data Collection Portal → Portfolio → Asset List → edit → Meters & Consumption → Meter consumption
How it works
- Identifies meters with missing consumption data
- Uses historical trends to project likely values
- Populates suggested values directly in the entry view
- Allows users to review, edit, or accept before saving
Example:
If a meter’s last 3 years show a consistent trend, the AI forecast auto-fills next year’s expected consumption for validation.
Availability
- Energy
- Water
- Waste
Bill scraping and verification
Purpose
To automate invoice data extraction and reduce manual data entry.
Location
Data Collection Portal → Portfolio → Asset List → edit → Meters & Consumption → Meter consumption → Paperclip icon
Process
- Upload a PDF utility invoice at a meter consumption entry
- AI extracts key fields (meter ID, billing period, consumption, cost)
- Extracted values populate the relevant fields
- Optionally connect to Scaler’s verification partners
Advantages
- Saves time
- Reduces transcription errors
- Supports consistent, auditable data entry
- Integrates with manual and API-based workflows
Security and privacy
- All AI tools operate entirely within Scaler’s AWS environment
- No external AI APIs are used
- Data remains regionalised and compliant with GDPR, SOC 2, ISAE 3000, and ISO 27001
Best practices
- Review AI-suggested forecasts and measures before approval
- Use AI data quality reports regularly to monitor anomalies
- Use Lumi for quick internal queries and orientation
- Enable verification workflows for audit-ready portfolios