Three in five of your competitors install the same brand.
That is not your differentiator.

We run a directory of geothermal installers. There are 2,355 published contractors in it across North America, and until this week nobody had properly looked at what they say about themselves.

Two things came out of it. One is a number large enough to change how you position a geothermal business. The other is an absence so complete it took a second investigation to believe.

One brand owns this trade

122 distinct equipment brands are named across the directory. One of them accounts for more than the next four combined.

Shares add to more than 100% because installers can name several brands — though most do not. The median installer names exactly one.

Separately from what installers say, 929 of the 2,355 are recorded as registered WaterFurnace dealers — 39.4% of the entire directory, whether or not they list a brand at all.

The marketing line this kills

“We install WaterFurnace” is on a great many geothermal websites. As a reason to choose one contractor over another, it is worth approximately nothing, because three in five of the contractors a customer might call say the same thing.

This is not an argument against carrying the brand. It is a good product and there are sound reasons the trade has settled on it. It is an argument against leading with it. A differentiator that 953 of your competitors also claim is not a differentiator; it is table stakes wearing a badge.

The same goes for most of what fills a geothermal homepage: efficiency percentages, tax credit explainers, how a ground loop works. All true, all useful, all identical to everyone else’s.

What the trade does not publish

So we looked at the opposite — the fields where a contractor could say something specific, and found them mostly empty.

A homeowner is contemplating a job that will run well into five figures, with a payback measured in years. Before they call anybody, the trade tells them: nothing about depth, nothing about cost, and in two thirds of cases nothing about which loop types are even on offer.

So we checked their own websites

An honest objection: perhaps those fields are empty because our import never captured them, not because contractors are silent. If so, this whole finding is about our data collection and nothing else.

So we tested it. We took 30 listings that have a website and whose loop-type field is empty in our directory, and read what those 30 businesses publish on their own sites.

Three of the thirty would not load at all. Two fail the security handshake outright — a visitor gets a browser warning rather than a website — and one has no working domain. Two more returned pages our reader could not see, and we excluded them rather than count them as silent.

Of the 25 sites we could genuinely read: 3 state a loop type. 1 states a bore depth. None states a cost per foot. 2 mention IGSHPA. 3 show a licence number.

That settles it, and it settles it in the opposite direction to the objection. 12% of these contractors state a loop type on their own website, against 35% across our directory as a whole. The listings are better populated than the websites. The silence belongs to the trade.

Half of them never say the word

One thing we were not looking for turned out to be the most striking of all.

11 of those 25 websites never mention geothermal at all. Not on the homepage, not on any page the homepage links as geothermal.

These are not empty sites. One runs to more than 12,000 words of text. They are heating, cooling and plumbing companies that appear in a geothermal directory, presumably because they do the work, and never raise the subject with anyone who visits them.

Consider what that means for a homeowner researching geothermal. They find the company, they click, and there is no evidence the company does the thing they were searching for. Whatever was spent to get them to that page is spent.

What to do instead

Everything below is free, takes an afternoon, and almost nobody in this trade has done it.

Say the word. If you install geothermal, say so on the page people land on. Nearly half the sample does not. This is the lowest bar in the whole piece and half the trade is under it.

Publish your loop types. Horizontal, vertical, pond, open loop — whichever you actually do. Doing this puts you ahead of 65% of the directory and, on the evidence of their own websites, closer to 88% of your competitors.

Publish your licence number. 91% do not. It costs nothing and it is checkable, which is exactly what makes it worth more than an adjective.

If you hold IGSHPA certification, lead with it. 7% of this trade does. That is a genuinely scarce credential and most of the people who hold it are burying it.

Give a real number about depth or cost. A range, with the conditions attached. Nobody — not one contractor in 2,355 — is doing this. The first one to do it honestly will be the only page a homeowner finds that answers the question they actually have.

Notice what is not on that list: a rebuild, a campaign, or a better brand. The differentiator is not the equipment, because the equipment is the same. It is being the one contractor who answers a question.

What we are not claiming

We would rather draw the lines ourselves than have you find them.

We cannot verify where the brand data originally came from. We can confirm what each listing says. We cannot confirm the source it was collected from, and we are not going to imply otherwise.

Thirty websites is indicative, not conclusive. It decisively answers whether our import is the problem, because the result is so lopsided. It does not support a precise percentage about the whole trade, and we have not stated one.

We read the homepage and at most one linked page. A loop type buried three clicks deep would be missed. Read every website figure as “publishes this where a customer would readily see it”.

A blank field is a blank field. A contractor may know their bore depths perfectly well and simply never write them down. That is rather the point.

How we measured this

All directory counts are read-only queries against the GeothermalFinder database on 4 August 2026, covering all 2,355 published listings in North America. Brand figures are computed by parsing each listing’s brand list properly, not by text matching — 498 listings hold an empty list, which is not the same as naming no brands and is counted as such. The 60.6% describes the 1,573 listings that name at least one brand, not all 2,357.

The website sample was 30 listings drawn at a fixed interval by record number, so the selection is reproducible rather than chosen. Of those, 3 could not be loaded, 2 returned no readable text and were excluded, and 25 were read.

This is directory measurement, not client results. We have no client results in this trade to report, and we say so on the evidence page.

Published by Groundline Marketing, 4 August 2026. Corrected the same day.