IESO Commercial Reconciliation & Settlement Intelligence

ReconIntel is A18 Analytics’ commercial reconciliation intelligence platform for IESO market participants. It is designed to help utilities, generators, traders, wholesale participants, and large energy users reconcile settlement outcomes, investigate variances, monitor financial exposure, and prepare evidence-backed responses before settlement issues become commercial or compliance risks.

At its core, ReconIntel brings together public IESO market data, confidential participant reports, internal shadow settlement data, charge logic, rule references, and evidence workflows into one governed reconciliation environment. Rather than treating settlement review as a manual spreadsheet exercise, ReconIntel turns it into a structured, auditable, and decision-ready process.

The platform supports settlement analysts, finance teams, compliance leads, and operational stakeholders by helping them answer the questions that matter most: what changed, why it changed, which charges caused the variance, whether the issue is material, what evidence supports the claim, and whether a Notice of Disagreement should be prepared.

ReconIntel is especially relevant as the Ontario market transitions into a more complex post-Market Renewal environment. With Day-Ahead Market structures, Real-Time Market settlement, LMP and nodal pricing considerations, and participant-specific settlement logic becoming more important, market participants need more than a dashboard. They need a controlled reconciliation workbench that connects data, rules, evidence, and action.

What ReconIntel Helps With

ReconIntel supports:

IESO settlement statement reconciliation
Shadow settlement comparison
Public and confidential report ingestion
Day-Ahead and Real-Time market data analysis
LMP and nodal pricing readiness
Exception and variance investigation
Evidence package generation
NOD readiness assessment
Settlement intelligence dashboards
Governance, lineage, and auditability

Who It Is Built For

ReconIntel is designed for:

Local Distribution Companies
Generators and storage operators
Traders and wholesale market participants
Class A loads and large industrial energy users
Settlement analysts
Finance and revenue assurance teams
Compliance and regulatory teams
Energy consultants supporting market participants

Why It Matters

Commercial settlement is not just an accounting activity. It is a control environment. A single unexplained variance can affect cash flow, dispute readiness, regulatory confidence, and executive decision-making. ReconIntel helps market participants move from reactive settlement review to proactive settlement intelligence.

Settlement Intelligence for the IESO Market

Electricity markets are not only physical systems. They are also commercial systems, and every megawatt that moves through the grid eventually becomes a financial line, a charge code, a settlement statement, a variance, or a question that somebody must explain. That is the problem ReconIntel was designed to address.

In the Ontario IESO market, participants are not simply trying to understand whether they consumed, generated, traded, or delivered energy. They are also trying to understand how those activities were settled, whether the settlement result agrees with internal expectations, whether public market data supports the outcome, whether confidential IESO reports align with internal records, and whether any variance is large enough to justify further review.

In many organizations, this work still lives across spreadsheets, downloaded reports, email threads, analyst notes, manual comparisons, and institutional memory. The result is that settlement review becomes slower than it should be, harder to audit than it needs to be, and more dependent on a few people who understand how the moving parts fit together.

ReconIntel takes a different view. It treats settlement reconciliation as a governed intelligence workflow, not as a one-off reporting task.

From Settlement Review to Settlement Intelligence

A normal dashboard can tell a team what the final number is. ReconIntel is designed to explain how that number was created. That distinction matters, because a settlement analyst does not only need to know that a variance exists. They need to know whether the variance came from price, quantity, charge logic, market data, metering quality, statement version changes, Day-Ahead versus Real-Time differences, Global Adjustment treatment, LMP assumptions, or a data-quality issue. In the same way, a finance lead does not only need to see the total bill. They need to know which portion of the settlement is energy, which portion is GA, which portion is uplift, which part changed from PSS to FSS, and which items may require commercial action.

ReconIntel is therefore designed as a reconciliation workbench. It connects ingestion, validation, settlement comparison, exception handling, evidence packaging, and business-facing visualization into one controlled product experience.

In practical terms, that means the user can move from:

IESO report received
→ file ingested
→ data staged
→ reconciliation run created
→ variance detected
→ exception opened
→ evidence reviewed
→ NOD readiness assessed
→ management view updated

without having to manually stitch the entire process together every time.

Why Market Renewal Makes This More Important

The Ontario market is becoming more complex, not less complex.

As Market Renewal introduces more advanced Day-Ahead and Real-Time structures, and as LMP and nodal pricing become increasingly important to certain participant types, settlement intelligence can no longer depend only on traditional total-bill review. Generators, traders, storage operators, LDCs, and Class A loads do not all experience settlement risk in the same way. Their formulas, drivers, KPIs, and decision points are different.

A generator may care deeply about resource-level revenue, nodal prices, operating reserve, and Day-Ahead versus Real-Time exposure. An LDC may care more about wholesale settlement, loss factors, GA pass-through, customer class mix, and metering integrity. A trader or wholesale participant may care about book-versus-settlement differences, DA/RT positions, congestion components, and allocation factors. A Class A load may care most about GA exposure, coincident peak contribution, peak-shaving impact, and Class A versus Class B counterfactuals.

This means ReconIntel cannot be a one-size-fits-all calculator, and Instead, the product is being structured around a simple but powerful idea:

The plan determines the depth of platform access.
The participant profile determines the settlement intelligence logic.

That allows ReconIntel to preserve one clean commercial product structure while still adapting the math, dashboards, evidence templates, and workflows to the user’s actual market role.

The ReconIntel Workflow

The product experience is built around a few core workflows.

First, ReconIntel brings data into a controlled environment. This includes public IESO reports, confidential participant files, internal shadow settlement data, and supporting market data. The point is not merely to store files. The point is to make them traceable, validated, and usable for reconciliation.

Second, the system stages and validates the data. This matters because settlement analysis is only as trustworthy as the data feeding it. Missing intervals, estimated values, VEE flags, mismatched charge lines, stale public feeds, or incomplete statement versions must be visible before a user starts making decisions from the outputs.

Third, ReconIntel runs the reconciliation logic. It compares internal or shadow expectations against IESO statement data and produces variance outputs. These variances are then grouped into exceptions, classified by likely driver, and prioritized by materiality.

Fourth, the platform creates evidence. This is where the product moves beyond ordinary analytics. A variance is useful only when it can be defended. ReconIntel is designed to trace material exceptions back to source files, charge lines, statement versions, rules, formulas, market data, and lineage references.

Finally, ReconIntel supports action. That may mean preparing a Notice of Disagreement, escalating a settlement issue, briefing finance leadership, or simply closing an exception as immaterial. In every case, the workflow is structured so that the user can see not only the number, but the evidence behind the number.

Settlement Intelligence Cockpit

One of the most important modules in ReconIntel is the Settlement Intelligence Cockpit, and this is not meant to be a decorative dashboard, but a decision-support layer of the platform.

The cockpit is designed to answer questions such as:

What is our net settlement position?
Which charge categories drove the outcome?
How much variance exists between shadow and IESO?
Which lines are above dispute threshold?
What changed between PSS, FSS, and RCSS?
What is our NOD exposure?
How much time remains before the deadline?
What data quality issues may have caused the problem?

For finance users, it provides a clearer view of exposure, credits, debits, charge categories, and cash-flow implications. For settlement analysts, it highlights variance drivers, charge-line issues, and source-level evidence. For compliance teams, it tracks deadline risk, dispute readiness, and documentation completeness. For operations teams, it surfaces data integrity, VEE, metering, and estimated-data concerns.

This is where the platform becomes more than a reconciliation engine. It becomes an operating layer for commercial settlement control.

Evidence, Governance, and Trust

ReconIntel’s strongest design principle is traceability, because every important number should be able to answer the question:

Where did this come from?

That means settlement figures should link back to source files. Variances should link back to charge lines. Charge lines should link back to rule or equation references where available. Evidence packages should link back to the staging rows, ingest runs, and source batches that produced them. If a user is preparing a dispute, they should not have to rebuild the evidence manually after discovering the issue.

This is why governance is not being treated as a back-office feature. In ReconIntel, governance is part of the user experience. The platform is meant to show data lineage, quality status, source file history, parser status, validation outcomes, and audit events directly in the product.

In settlement environments, trust is not created by nice visuals. Trust is created by explainable numbers.

AI in ReconIntel

The AI layer in ReconIntel is deliberately controlled.

The deterministic engine calculates the settlement comparison. AI does not replace the math, override the rules engine, or submit disputes automatically. Instead, AI or template-assisted intelligence helps explain the result, classify the likely cause, summarize evidence gaps, and draft language for analyst review.

This is the correct role for AI in a regulated commercial workflow.

For example, if the system identifies a material variance on a generator resource, the AI layer can help explain whether the issue appears related to a nodal price mismatch, a Day-Ahead versus Real-Time spread, a missing interval, estimated data, or a resource-to-node mapping problem. If the evidence package is complete, it can also help prepare a draft explanation for a Notice of Disagreement.

However, the final decision remains with the human analyst, and that balance is important, as ReconIntel should accelerate professional judgment, not replace it.

How ReconIntel Fits A18’s Methodology

ReconIntel reflects A18 Analytics’ broader approach to governed decision intelligence.

The product does not treat analytics as a final dashboard sitting on top of messy processes. Instead, it treats data ingestion, validation, metadata, business rules, reconciliation logic, evidence, workflow, user experience, and governance as one connected system.

That is the A18 methodology in practice.

The goal is not simply to visualize settlement data. The goal is to turn settlement data into a controlled decision environment where analysts, finance teams, compliance teams, and executives can understand what happened, why it happened, what is at risk, and what action should follow.

In that sense, ReconIntel is both a product and a demonstration of A18’s philosophy: enterprise analytics should be governed, explainable, operationally useful, and designed around the decisions people actually need to make

ReconIntel: Building Settlement Intelligence for the IESO Market