Skip to content
GridVision AI

Interconnection decision intelligence

The Interconnection Decision Expert for Energy Developers

Know where to invest, what to advance, and when to walk away.

GridVision turns fragmented interconnection evidence into executive-ready recommendations across markets, sites, queue positions, and portfolios—before the next study dollar, engineering commitment, or development investment is made.

Bring One Live DecisionSee How GridVision Works
  • INTERCONNECTION QUEUE
  • PJM
  • QUEUE POSITION
  • MISO
  • STUDY CAPITAL
  • CAISO
  • PORTFOLIO PRIORITIZATION
  • NYISO
  • EVIDENCE LINEAGE
  • ISO-NE
  • DEVELOPMENT DECISIONS
  • SPP
  • DECISION RECORD

The executive problem

The hard question is not finding the data. It is deciding where engineering time and development capital should go.

Queue, market, site, and project evidence arrive fragmented and in different states of certainty. Management still has to commit the next study dollar. GridVision turns that evidence into a decision-ready recommendation with its basis attached.

Today that evidence sits in

  • ISO records and project histories
  • Consultant and analyst workstreams
  • Engineering and development assumptions
  • Portfolio spreadsheets and committee materials

Point of view

GridVision helps energy developers decide where to invest, what to advance, and when to walk away.

A decision system, not a data aggregator. The queue is not merely a filing problem—it is a capital-allocation problem, and every position must keep earning its next commitment.

Public and authorized evidence
Analysis is bound to project, market, and queue evidence the team is authorized to use.
Deterministic analysis
Screening is rule-based and repeatable, not a generative guess about the network.
Transparent provenance
Source, as-of time, fetch time, and dataset or model version are retained where present.
Explicit unknowns
Verified evidence, user inputs, assumptions, and missing data stay separately labelled.
Executive-ready recommendation
The output is a decision structured for management review, with its evidence attached.

Market reality

Most capacity that enters an interconnection queue never gets built.

Published industry research, not GridVision results. These figures describe the market developers are deciding in.

13%

of capacity that requested interconnection between 2000 and 2020 had reached commercial operation by the end of 2025

75%

of that same capacity was withdrawn from the queue

~2,060 GW

of generation and storage was still actively seeking interconnection at the end of 2025

5+ years

median time from interconnection request to commercial operation, for projects built in 2025

Source: Lawrence Berkeley National Laboratory, Queued Up: 2026 Edition. Data as of year-end 2025; published June 2026.

Six ISO markets represented

PJM, MISO, CAISO, NYISO, ISO-NE, SPP. Represented sources — coverage is not universal across ISO/RTO markets.

Read the public record for each

Three evidence-gated verdicts

PURSUE, REVIEW, DEPRIORITIZE express relative development priority, not an investment decision.

N/A rather than a guess

A missing source field is shown as N/A and never replaced with sample or inferred data.

How GridVision works

Fragmented evidence becomes one accountable recommendation.

  1. 0.1

    Intake

    Define the project, market, POI, decision owner, and evidence cut-off in a structured intake. No input is guessed.

  2. 0.2

    Evidence

    Bind the analysis to authorized project, queue, and network evidence. Gaps are recorded rather than filled.

  3. 0.3

    Engineering screening

    Run deterministic DC screening when an authorized current network model and required ratings are available.

  4. 0.4

    Decision policy

    Apply the evidence gate. Where required evidence is absent, the policy forces review rather than ranking.

  5. 0.5

    Executive recommendation

    Persist an append-only Decision Record with verdict, evidence coverage, risks, unknowns, and next diligence.

GridVision owns the deliverable. Management review and formal utility and engineering studies remain authoritative throughout.

Product capabilities

What is available now, what depends on evidence, and what is not built yet.

Each capability carries its own state and scope. Nothing on this page is described as live unless it is.

  • Available now

    Queue Explorer

    Search represented projects and locations across the national interface and open a project detail page.

    Covers represented sources only; coverage is not universal across ISO/RTO markets.

  • Available now

    Structured project intake

    Define a project or comparison in a structured intake across supported market and project fields.

    No field is inferred. Unsupplied inputs remain unsupplied.

  • Available with required evidence

    Project and queue evidence

    Use a verified current queue record when a supported provider returns one for the project in question.

    If the record is unavailable, the analysis does not fall back to sample evidence.

  • Available with required evidence

    DCPF / PTDF / LODF screening

    Deterministic DC screening over an authorized current network model with the required ratings and injections.

    Not an AC, stability, short-circuit, protection, or utility study. Unavailable without an authorized snapshot.

  • Available with required evidence

    Hosting-capacity screening

    Screening support within the same deterministic DC scope as the rest of the network analysis.

    Scope is limited to the authorized network model supplied; it is not a utility hosting-capacity publication.

  • Available now

    Evidence lineage

    Retain source, as-of time, fetch time, dataset and model versions, hashes, freshness, and transformation notes where present.

    A lineage reference records provenance. It is not independent verification of the source.

  • Available now

    Decision Record

    Persist an append-only, tenant-scoped record of the analysis that authorized users can reopen later.

    A new analysis creates a new record. Existing records are never rewritten.

  • Available now

    Executive report export

    Export an evidence-backed JSON or PDF report of the Decision Record for a management meeting.

    Available to roles holding export permission. Export is audit logged.

  • Available with required evidence

    Scenario comparison

    Compare options manually with explicit input references and visible source limitations.

    Sample or unknown queue evidence is not decision-eligible in a comparison.

  • Not available — planned

    In-product review and sign-off

    Capture of decision owner, reviewer, comments, and approval inside the product.

    Not shipped. Management approval is recorded in the customer's own system of record.

  • Not available — planned

    Forecast provenance register

    A decision-native register comparing forecast versions, assumptions, owners, conflicts, and decision impact.

    Not shipped. Forecasts may be recorded as evidence inputs, but are not independently verified.

  • Not available — planned

    Decision-to-outcome tracking

    Following a formal Decision Record through to its later real-world result as one chain.

    Not shipped. Do not rely on closed-loop outcome tracking.

Illustrative decision format

One accountable recommendation. One management action.

Decision Record — sample format

Illustrative — sample data

Recommendation

REPRIORITIZE

Decision basis

Stronger competing position, unresolved study exposure, and higher near-term diligence burden.

Management action

Hold the next capital commitment until the decisive diligence item is resolved.

Alternative

Advance the competing position with the stronger current evidence case.

Decisive diligence

Resolve the study exposure that can change the portfolio order.

Evidence trail

Attach represented project facts, assumptions, N/A fields, and source references.

Evidence coverage

Which inputs were verified, which were user-supplied, and which remain assumptions.

Unknowns

Fields with no available source stay listed as unknown and are shown as N/A.

Format example only — not a customer case or completed analysis. Evidence coverage is not a success probability, and a verdict expresses relative development priority, not an investment decision.

Public-record retrospectives

Read the queue record for yourself.

Readings of a public dataset, not customer case studies. Every row can be checked against the source.

All retrospectives

Trust and evidence

Every recommendation carries its supporting evidence.

Decision evidence across six major ISO markets—PJM, MISO, CAISO, NYISO, ISO-NE, and SPP—organized for management review. Represented project records carry source attribution, and absent source fields are shown as N/A rather than filled with invented values.

Source attribution
Each evidence reference carries the source it came from.
Fetch and as-of time
When the evidence was published and when GridVision retrieved it are both retained.
Model and dataset version
The network model, dataset version, and transformation notes travel with the analysis.
Missing data remains missing
An unavailable source field is shown as N/A. It is never filled with an inferred value.
Sample data cannot become authoritative
Sample or unknown evidence is not decision-eligible and cannot promote itself into a record.
AI explains, it does not create facts
Assistance can explain a Decision Record. It cannot generate engineering facts or supply missing evidence.

Founding design partner

A bounded engagement on one real decision.

GridVision applies its decision process to one real, customer-authorized development decision over 30–45 days, and the team judges whether the resulting evidence structure and decision record improve management review.

Decisions in scope

  • One project at a defined point of interconnection
  • Two candidate sites or interconnection pathways
  • Two competing projects
  • One historical decision, reviewed retrospectively

What the partner provides

  • One executive sponsor and one project or data owner
  • A clearly defined decision and decision deadline
  • Data the organization has the right to authorize for the pilot
  • Confirmation of the applicable tenant and access boundary
  • Participation in the evidence review and final management review

Deliverables

  • 0.1

    Evidence and provenance map

  • 0.2

    Data-readiness and evidence-gap assessment

  • 0.3

    Facts, assumptions and unknowns register

  • 0.4

    Supported alternative-pathway comparison

  • 0.5

    Recommended next-diligence plan

  • 0.6

    Management decision record

Where required evidence is missing, GridVision identifies the gap rather than substituting for it — an evidence-gap assessment remains a valid outcome, but an unsupported model-based conclusion is not issued. Success does not require a predetermined recommendation, model result, or commercial outcome. Formal utility determinations and engineering studies remain authoritative, and final judgment and action remain with the customer's management.

  • Renewable developers
  • Storage developers
  • Data-center developers
  • Infrastructure developers
  • Transmission and interconnection leaders
  • Development and portfolio executives

Next step

Bring one live interconnection decision.

Bring a real project, market, or queue decision your team is weighing. GridVision will prepare the recommendation for management review, with its evidence and unknowns attached.

Scope is agreed together: the project, decision owner, evidence cut-off, and which sources you are authorized to use.

Submit one decision for management-review scoping.

By submitting this form you agree to our Terms of Service and acknowledge our Privacy Policy.