What Public Grid Data Can't Tell You About Interconnection

A free national map now scores roughly 66,000 US neighborhoods on which kind of distributed energy resource the local grid actually calls for. It is the most useful piece of public grid data to reach distribution scale in years, and it is going to be misread. GridSolver tells you where a distributed resource would be worth the most. It does not tell you whether you can connect one there.

That gap is the entire story, and it is not a footnote. A location can score at the top of GridSolver’s value ranking and still be interconnection-infeasible on the day an engineer opens the circuit model. If you are a municipal utility or a cooperative planning around this tool, the difference between those two statements is the difference between a defensible siting argument and a study you pay for twice.

The tool came out of the OpenEAC Alliance on June 22, 2026, built by WattCarbon with co-builder Resilience Energy. It is live and public at gridsolver.wattcarbon.com. What follows is what it measures, what it does not, and where it fits in a procurement conversation.

What GridSolver actually measures

The inputs are all public, which is the first thing worth saying plainly. GridSolver pulls hourly net load from 53 balancing authorities, day-ahead congestion prices from seven US wholesale markets, and building footprints from NREL’s End-Use Load Shapes. Market feeds refresh daily. The national map re-scores monthly. The full methodology is published.

From those inputs it does two things. It measures how sharply an area’s demand concentrates into a small number of hours, and it identifies in dollar terms where moving power past a transmission constraint is most expensive. Then it classifies each location into one of four distributed resource categories based on the pattern it sees.

WattCarbon CEO McGee Young framed the point of the exercise this way: “For too long, the value of a DER has been a matter of opinion, and opinion is a terrible foundation for a market.” That is a fair description of the problem being solved. It is also a precise description of the scope, because value is what the tool models. Feasibility is not.

A DER value map is not a hosting capacity map

This is the confusion to head off before it costs someone a study cycle.

A hosting capacity map answers an engineering question: given this circuit’s thermal limits, voltage profile, protection scheme and backfeed behavior, how much generation can connect at this point without a system upgrade? It is built from the utility’s own circuit model and it is the thing an interconnection application gets measured against.

A DER value map answers an economic question: given how load and congestion price behave in this area, what kind of resource would be worth the most here? GridSolver does not model thermal limits. It does not model voltage. It does not model protection coordination or backfeed. Those four things decide interconnection, and none of them are in the data set.

Both maps can be right and point in opposite directions. A feeder can sit in a high-congestion pocket that makes storage extremely valuable and simultaneously have no headroom to accept it without a reconductor. Nothing in the value score anticipates that, and nothing in it should be expected to. The failure mode is not the tool being wrong. It is a reader treating an economic screen as an engineering answer.

Our own coverage of DERMS moving from pilot to mainstream procurement is the other half of this picture. ComEd manages roughly 8 GW of distributed resources on its DERMS and targets 650 MW of flexible interconnection by 2031, using managed curtailment to fit more solar and storage onto existing circuits rather than rebuilding them. That is a utility deciding feasibility with its own circuit model and its own control layer. A public value map cannot substitute for it, and the utilities furthest along on this are the ones least likely to confuse the two.

Neighborhood level grid data does not follow feeder topology

The scoring unit is a neighborhood. That word is doing real work and it is worth being exact about, because our own research file got this wrong for six weeks before we corrected it.

A neighborhood is not a substation service area and it is not a feeder. One substation typically serves many neighborhoods, and neighborhood boundaries follow streets and census geography rather than the way conductor actually runs. Two houses on opposite sides of the same street can sit on different feeders fed from different stations. A score attached to the polygon between them is an average across an electrical boundary the polygon cannot see.

That does not make the number useless. It makes it a screening layer. It tells you which parts of a territory are worth a closer look, in the same way a load-growth heat map tells you where to point a planning study. It does not tell you what the study will find.

The real contribution is grid data transparency at national scope

There is a version of this story that overstates the novelty, and it is worth refusing.

Utilities have had visibility into their own circuits for a long time. Several states already require published hosting capacity maps, California, New York and Minnesota among them. Neighborhood-resolution grid analysis is not new, and anyone claiming this kind of visibility did not previously exist is going to get corrected by the first distribution engineer who reads it.

What genuinely did not exist is a free, national, methodologically uniform layer of public grid data. State hosting capacity maps are utility-specific. They are defined differently, updated on different cadences, and are not comparable across territories. Comparing a New York utility’s map to a Minnesota utility’s map tells you more about the two filing requirements than about the two grids.

GridSolver’s contribution is comparability at national scope, plus a published method you can audit. For anyone doing market analysis across territories rather than engineering inside one, that is the more valuable property, and it is the one to cite. Grid data transparency at national scale is a different achievement from grid data resolution, and the two get conflated constantly.

Four categories, four different equipment answers

The classification scheme is the part a distributor or a small utility can borrow directly, because each category implies a different bill of materials.

Solar and efficiency flags areas with flat demand. The equipment conversation there is inverters and medium-voltage interconnection.

HVAC flexibility flags winter and summer stress zones. That is a controls and demand response conversation before it is a hardware conversation.

Non-HVAC flexibility flags shoulder-season congestion, which is the least intuitive of the four and usually the most process-dependent.

Load shifting flags year-round recurring stress. That is where storage earns its place, and where the virtual power plant procurement wave is already pulling equipment orders.

Characterizing a service territory as HVAC-flexibility-shaped rather than load-shifting-shaped is a genuinely useful sentence for a small utility that has never framed its DER opportunity in equipment terms. It is the kind of segmentation that used to arrive attached to a consulting invoice.

The commercial tell

One paragraph on why this tool exists, because it changes how you read it.

Resilience Energy installs distributed resources on residential property and needs a defensible valuation method to turn those installs into a bankable PPA. CEO Ameet Konkar said as much: “What’s missing is a way to turn this new capacity into a PPA, similar to how large companies purchase renewable energy.” GridSolver is that valuation method made public. Open-sourcing your number is how you get a counterparty to accept it.

That is not a criticism. A published methodology with public inputs is a higher citation standard than most vendor analysis clears. It does mean the tool is built to be persuasive in front of utilities and offtakers, and the signal to watch is whether a utility formally engages with these numbers. No utility partners are named today. The first utility to either adopt or publicly contest a GridSolver score is the moment it stops being a marketing artifact and becomes a reference.

How to use DER siting data without getting burned

Three rules cover most of the exposure.

Pair it with the utility’s actual hosting capacity data before any siting decision. The value score narrows the search. The circuit model closes it. If your utility publishes a hosting capacity map, the two together are a real screening workflow. If it does not, the value score is a reason to ask, not an answer.

Check the vintage against your question. Market feeds refresh daily and the national map re-scores monthly, which is fast for public grid data and slow relative to a load that changed last quarter. A distributed load arriving behind a residential service transformer does not show up in a monthly re-score until it is already there.

Do not cite it as a substitute for an interconnection study in front of a board or a commission. State what it is: a national, uniform, published-method screen of where distributed capacity would be worth the most. That claim is defensible and it is strong enough on its own. The overstated version is the one that gets challenged.

Public grid data is getting better fast, and most of the useful pieces are free if you know where they are and what each one can carry. The Feeder is our free monthly briefing on what changed and what it means for equipment timing. Sign up here.

Related Reading

• DERMS Grid Modernization Procurement Goes Mainstream
• The VPP Procurement Wave Is Here
• The Distributed Data Center Lands on a Service Transformer