How we select builders
Five builders per market, chosen against thirteen published criteria, reviewed at least annually. No position is for sale. This page is the promise the rest of the site is published under.
We select five builders per market by verifying licensing and warranty standing first, then assessing documented work against thirteen published criteria and writing a rationale for each position. There is no numerical score, because a weighted rubric would imply a measurement nobody took. Builders cannot buy inclusion or ranking, and paid placement is labelled separately wherever it appears.
Why only five builders per market
- A profile is only worth publishing if it is researched properly and kept current. Five per market is roughly what can be maintained to that standard; fifty per market is a database that rots.
- A list of two hundred builders is not a shortlist, it is a phone book. It transfers the entire selection problem back to the reader, which is the problem they came here to solve.
- A short list is falsifiable. If a selection is wrong, it is visibly wrong and it can be argued with. A long list can never be wrong because it never committed to anything.
- Scarcity has to be honest to be worth anything. The list is short because research capacity is finite, not because inclusion is being rationed to create demand for it.
What the list is not
- It is not a claim that these are the five best builders in the market. It is a claim that these five were researched against published criteria and that the research is on the page.
- It is not a recommendation or an endorsement, and it does not replace your own due diligence on licensing, insurance, warranty, references, financial standing or contract terms.
- It is not a statement that an unlisted builder is inferior. A builder’s absence means we have not published research on them — most often because we have not researched them yet.
- It is not a guarantee of any outcome on your project. No third party can offer that, and anyone who does should be treated with suspicion.
What we assess, and what we do when the evidence is missing
The third column is the part most methodologies leave out, and it is the one that determines whether a published profile is honest. A criterion with no evidence is recorded as unassessed, never inferred.
| Criterion | What is assessed | Evidence | When the evidence does not exist |
|---|---|---|---|
| Portfolio quality and depth | Whether completed work shows consistent execution across multiple projects rather than one exceptional house, and whether the volume of documented work is sufficient to judge consistency at all. | Published project documentation, architectural and design-press coverage, photographer and architect credits, and site visits where access is offered. | A builder with little documented work is not penalised as poor — they are simply not assessable yet, and are held for a future review rather than ranked. |
| Relevance to the local luxury market | Whether the builder’s documented work is actually in the market and the segment the selection covers, rather than adjacent to it. | Project locations, stated service areas, and the municipalities the builder holds permit history in where that is public. | Where service area cannot be established from evidence, the field is published as unverified rather than inferred from marketing copy. |
| Demonstrated custom-building experience | Genuine custom work — homes designed for a specific client and site — as distinct from production or semi-custom building marketed as custom. | Project variety, evidence of architect and designer collaboration, and the degree of plan repetition visible across the portfolio. | Held rather than assumed. This is a distinction builders describe generously about themselves, so it is assessed from the work. |
| Architectural range | Whether the builder can execute across more than one architectural language, or is deliberately and competently specialised in one. | Documented projects across styles, and the technical demands each style implies. | Narrow range is not a negative. A builder who does one thing exceptionally is a legitimate selection, and the profile says so. |
| Transparency of process | Whether the builder will describe their contract form, allowance practice, change-order process and schedule commitments plainly when asked. | Published process material, and direct response to our factual verification request. | A builder who declines to describe their process is recorded as having declined. That is a fact about them and it is published. |
| Warranty and professional standing | Current licensing or registration where the province requires it, current new home warranty enrolment, and good standing with the applicable regulator. | Public regulator registers — HCRA, BC Housing, RBQ, Alberta builder licensing — and warranty provider confirmation. | This is the one criterion with a hard floor. A builder who cannot be confirmed as licensed and warranty-enrolled where their province requires it is not eligible for selection, regardless of anything else. |
| Project documentation | Whether individual projects can be described factually — location, style, completion year, architect, designer — rather than only shown as photographs. | Builder-supplied project data, architect and designer confirmation, and published credits. | Undocumented projects are not published. A gallery without attribution is not evidence and is not treated as any. |
| Geographic and regulatory expertise | Demonstrated familiarity with the specific approval environment the buyer will face — the municipality’s process, the architectural review committee, the county’s servicing requirements. | Project history in the relevant jurisdiction, and the specificity of the builder’s own answers about it. | Assessed from project history rather than from claims. |
| Client education | Whether the builder helps buyers understand the process, the costs and the risks — or relies on the buyer not asking. | Published educational material, the substance of the builder interview, and the candour of their answers about what goes wrong. | Absence of published material is not a negative on its own; many excellent builders do not publish. The interview covers it. |
| Industry recognition | Peer and association recognition, where it comes from a body with an assessment process rather than a paid entry. | Awarding body records, verified directly rather than from the builder’s own account. | Unverifiable awards are not published. This is the field most commonly overstated in the industry and it is treated accordingly. |
| Consistency of available evidence | Whether the builder’s own claims agree with independently checkable records — corporate registry, permit history, regulator status, published credits. | Cross-referencing all of the above. | A material inconsistency that the builder cannot explain is disqualifying, and the profile is withdrawn rather than annotated. |
| Verified public reputation | Whether patterns in publicly available client feedback are consistent with the rest of the picture — attending to recurring themes rather than to individual reviews. | Public sources, weighted for pattern rather than for volume or score. | No numerical review score is published or used, because unverified aggregate scores are trivially manipulated and we would be laundering them by republishing them. |
| Responsiveness to factual verification | Whether the builder responds to a written request to confirm the factual portion of their profile, and whether the response is accurate. | Our own verification correspondence. | Non-response is published as `Not yet verified` on the profile. It is not disqualifying, and it is not hidden. |
How a selection is made
In this order, every time. The market boundary is set before any builder is considered, so it cannot be drawn around a preferred outcome.
Market definition
The geographic boundary of the selection is set and published, including which surrounding municipalities count as part of the market. This happens before any builder is considered, so the boundary cannot be drawn around a preferred outcome.
Candidate identification
Candidates are assembled from public permit records, architectural and design press, association membership lists, architect and designer references, and applications received. Applying is one route in among several and confers no advantage.
Eligibility screening
Licensing, registration and warranty enrolment are verified against the applicable public registers. A builder who cannot be confirmed is removed from consideration at this stage regardless of the strength of their work.
Desk research
Documented projects, service areas, specialties, collaborators and public record are assembled, with the source recorded for each fact. Anything that cannot be sourced is not written down.
Assessment against the published criteria
Each candidate is assessed against the thirteen criteria above. The output is a written rationale, not a score. Where two candidates are close, the written rationale has to articulate the difference — which is a useful discipline, because sometimes it cannot.
Verification request
Shortlisted builders receive the factual portion of their draft profile and are asked to confirm, correct or decline. Editorial text is not sent for approval and is never negotiated.
Publication
Five builders are published per market, each with a ranking position, a written rationale for that position, a verification status and a last-verified date.
Review
Each market selection is reviewed at least annually, and sooner if a material fact changes — a licensing lapse, a change of ownership, a warranty enrolment issue, or a substantiated pattern of complaint. Positions can and do change on review.
Conflict-of-interest policy
This site is published by a licensed Alberta real estate associate. That is disclosed on every page rather than buried, because it is a relevant fact about who is making these judgements.
- Where the editor has a commercial relationship with a builder in any capacity, that relationship is disclosed on the builder’s profile.
- Sponsored placement is labelled on the card, in the comparison table and on the profile. It is never presented as an editorial position.
- Introductions between buyers and builders are made at the buyer’s request. Where any compensation arises from an introduction, it is disclosed before the introduction is made — not afterward.
- A builder’s decision to buy or decline sponsorship has no effect on their editorial position, and a sponsored builder is not protected from an unfavourable factual statement that is accurate.
- No builder receives advance notice of an editorial change affecting a competitor.
Update frequency
Every market selection is reviewed at least annually. A review is triggered sooner by any material change: a licensing lapse, a warranty enrolment issue, a change of ownership or leadership, a substantiated pattern of client complaint, or a correction that materially alters a published fact. Every page carries the date it was last reviewed.
If we get it wrong
We will. The corrections process is published, corrections are recorded with the date and the substance of the change rather than silently edited, and the correction link appears on every substantive page.
Questions about selection
Can a builder pay to be included or ranked higher?
No. Editorial inclusion and ranking position are not for sale, under any arrangement. Sponsored placement exists — enhanced photography, sponsored educational content, lead routing, video interviews, expanded galleries, commissioned market reports — is sold separately, is labelled wherever it appears, and never moves an editorial ranking.
Why is there no scoring system?
Because a weighted 100-point rubric would be false precision. Assigning “Portfolio quality: 17/20” implies a measurement that was not taken, and the weights would inevitably be chosen to produce a ranking that had already been decided. A published ordinal ranking with a written rationale for each position is a smaller claim and a true one. If we ever introduce scores, every category, weight, data source, missing-data rule, update cadence and review process will be published here first.
Does applying help?
It gets your information in front of us, and nothing more. Applications confer no advantage over builders identified independently through permit records, published projects and architect references. Most applications will not result in a profile.
How often are selections reviewed?
At least annually, and sooner if a material fact changes — a licensing lapse, a change of ownership, a warranty enrolment issue, or a substantiated pattern of complaint. Positions can and do change on review.
What disqualifies a builder?
Two things. First, failing the licensing and warranty floor: a builder who cannot be confirmed as licensed and warranty-enrolled where their province requires it is not eligible regardless of the quality of their work. Second, a material inconsistency between their claims and independently checkable records that they cannot explain — in which case the profile is withdrawn rather than annotated.
Do you accept a builder’s edits to their profile?
We accept corrections to facts, with evidence, and we welcome them. We do not accept edits to editorial text. The factual portion of a draft profile is sent to shortlisted builders for confirmation; the editorial assessment is not sent for approval and is not negotiated.
What happens if a builder refuses to participate?
Their verification status is published as “Not yet verified” and the profile shows which fields are unconfirmed. Non-participation is not disqualifying and it is not hidden — it is a fact about the builder and it is recorded on the page.
Selection is a starting point, not a substitute for your own diligence.
Verify licensing, warranty enrolment, insurance and references on any builder — including one you found here. The method is published in full.