Exact documents
Evidence must match the formulation, model, size and tested application.
Best Research Rating
Every point in our rating comes from a document or a listing you can open. Popularity and advertising are capped so they cannot decide the order, missing evidence scores zero rather than a penalty, and our own opinion sits beside the score as a labelled confidence level that never touches the arithmetic. No brand receives points merely for its name or country of origin.
Research rating
Scores describe the evidence a buyer can find, not laboratory performance we measured. A product with few points is undocumented rather than poor, and a product can move up the moment a document lands. Every line shows the fact behind it so the arithmetic can be checked or disputed.
| Group | Points | What it measures |
|---|---|---|
| Product documentation | 30 | What the maker publishes about the product, and whether it can be read. |
| Claim traceability | 20 | Whether a document can be attached to the pack you are buying. |
| What you get | 20 | Equipment in the box, material in the pack, and where you can buy it. |
| Price per ounce of product | 15 | Kit price divided by the stated ounces of foam, not by a claimed board foot. Where a maker publishes only a gross can weight, the record says so. |
| Owner feedback | 15 | Platform ratings and volume, capped so they cannot decide the order. |
| Line | Max | How the points are awarded |
|---|---|---|
| Technical data sheet | 12 | Published with a date, a revision and a model number, 12. Published without them, 6. Not published, 0. |
| Safety data sheet | 6 | Published on the product page, 6. Not published, 0. |
| Fire evidence | 12 | Both indices at a stated thickness with no internal conflict, 12. The same with a conflicting line on the sheet, 8. Incomplete numbers, 4. A claim with no numbers, 0. |
| Net contents per can | 4 | Published and the same everywhere, 4. Published but channels disagree, 2. Not published, 0. |
| Model identifiers | 4 | Document and listing agree, 4. One is missing or they disagree, 2. Several sellers disagree, 0. |
| Yield stated with its conditions | 4 | Layer, distance and speed given, 4. A number with partial conditions, 2. None, 0. |
| R-value at a documented thickness | 4 | Stated at a thickness and consistent, 4. Stated but with a conflicting figure elsewhere, 2. None, 0. |
| Density published | 4 | A density or specific gravity with its test method, 4. None, 0. This is the condition the rim-joist exception needs and almost nobody prints. |
| Equipment in the box | 8 | Gun, cleaner and protective equipment, 8. Partial, 5. Cans only, 0. |
| Material in the pack | 6 | Total ounces published, 6. Not published so it cannot be compared, 0. |
| Where it is listed | 6 | Own store and both marketplaces, 6. Two channels, 4. One, 2. This counts listings found on the read date, not whether every size was in stock; the record says where stock was out. |
| Cost per ounce of product | 15 | Points = 15 x 0.484 / (price per ounce), capped at 15 and rounded to one decimal. The benchmark of 0.484 dollars an ounce is fixed: it was the lowest figure in the set when the rubric was written on September 7, 2026, and holding it fixed means adding a product never moves another product’s score. Price per ounce is the kit price divided by the stated ounces of foam, so a kit that includes a gun and cleaner is compared on its whole price. Where net contents are not published this cannot be computed and scores 0. |
| Rating and volume | 15 | Points = 15 x (rating / 5) x sqrt(min(ratings, 600) / 600), rounded to one decimal, from the anchor listing on the date the product was scored. Volume saturates at 600 ratings so a large review base cannot buy a position, and the square root keeps a small base from scoring near zero while still counting for less. These are platform figures, not a sample we drew. |
Two rules shape the design. Missing evidence scores zero, never a negative, so a product with few points reads as undocumented rather than poor and can move up the day a document lands. Owner feedback is capped at 15 against 30 for documentation, so review volume and advertising cannot buy a position.
Where the numbers come from. Documentation and traceability lines are read from each maker's data sheet, safety data sheet and manual, and from the listing fields on Amazon and Walmart. Price is cost per ounce of product at the 24-can kit, from a dated snapshot. Feedback is the platform rating scaled by review volume, saturating at 600 ratings. Every awarded line on a product page shows the fact behind it.
| Kit | Product documentation | Claim traceability | What you get | Price per ounce of product | Owner feedback | Total | Confidence |
|---|---|---|---|---|---|---|---|
| Kraken Bond FastCoat | 24 / 30 | 14 / 20 | 20 / 20 | 12.8 / 15 | 10.1 / 15 | 80.9 | Watching |
| Akfix ThermCoat | 18 / 30 | 16 / 20 | 20 / 20 | 14.3 / 15 | 12.4 / 15 | 80.7 | Watching |
| Sprayman SprayCoat | 18 / 30 | 14 / 20 | 20 / 20 | 15 / 15 | 10.9 / 15 | 77.9 | Watching |
| Stanley F990P SuperCoat | 18 / 30 | 14 / 20 | 18 / 20 | 14.5 / 15 | 9 / 15 | 73.5 | Watching |
| DAP Wall & Cavity Foam (research set) | 24 / 30 | 18 / 20 | 10 / 20 | 6.7 / 15 | 7.8 / 15 | 66.5 | Watching |
| ixirpro Spray Foam Kit (research set) | 0 / 30 | 8 / 20 | 20 / 20 | 15 / 15 | 8.2 / 15 | 51.2 | Open |
| Adhesive Guru QuickCoat (research set) | 0 / 30 | 8 / 20 | 20 / 20 | 14.3 / 15 | 5.2 / 15 | 47.5 | Open |
| Vega Bond SF001 Purplecoat (research set) | 0 / 30 | 8 / 20 | 17 / 20 | 8.3 / 15 | 12.5 / 15 | 45.8 | Open |
| Y’allCanDIY Spray Foam Insulation Kit (research set) | 0 / 30 | 10 / 20 | 20 / 20 | 10 / 15 | 4.7 / 15 | 44.7 | Open |
| BIGRIZY GUARD Spray Foam Insulation Kit (research set) | 0 / 30 | 8 / 20 | 18 / 20 | 14.8 / 15 | 3.3 / 15 | 44.1 | Open |
| BONDURE Spray Foam Insulation Kit (research set) | 0 / 30 | 10 / 20 | 18 / 20 | 13.9 / 15 | 1.8 / 15 | 43.7 | Open |
| Magic Chems Spray Foam Kit (research set) | 0 / 30 | 6 / 20 | 15 / 20 | 14.5 / 15 | 5.2 / 15 | 40.7 | Open |
| BEEST FullStop | 4 / 30 | 4 / 20 | 14 / 20 | 0 / 15 | 12.9 / 15 | 34.9 | Open |
| TYTAN Professional Thermospray (research set) | 0 / 30 | 8 / 20 | 10 / 20 | 10.8 / 15 | 1.1 / 15 | 29.9 | Open |
Each kit name links to its scored lines. The full comparison, with the reason behind every point, is on the kit ranking page.
Score and confidence
The score is what the documents support. The confidence level is what we think of that evidence, stated as opinion and kept out of the arithmetic on purpose, so that no view of ours can move a product up or down the order. Today nothing is settled, which is the level doing its job rather than a verdict on the category.
| Level | What it means |
|---|---|
| Settled | The documents agree with each other and with the listings, and no question is outstanding. |
| Watching | The documents are there and we have asked the seller something we have not had an answer to. |
| Open | Core documents are missing or a headline claim cannot be checked at all. |
The score answers "what do the documents support?" The confidence level answers "how settled is our own record?" Keeping them apart means a reader can dispute either one on its own terms: the score by pointing at a document, the confidence level by pointing at a question we have left open. A short desk note beside each product says which question that is.
Evidence must match the formulation, model, size and tested application.
Videos are observations under the conditions shown, not a replacement for field testing. Brand demonstrations and sponsored creators are identified separately from independent evidence.
R-value stays attached to stated thickness and evidence; quantities, pack size and price use transparent units.
Verified, manufacturer-stated, under review, pending and unsupported claims remain visibly separate.
Advertising reach, sales signals and owner feedback do not substantiate a fire claim. Each product needs traceable evidence for the exact claim and configuration. An unresolved report does not establish a failed test or another fire classification.
Evidence contract
Read three fields together: source type, product identity and the status of the particular claim.
| Field | Meaning | Limit |
|---|---|---|
| Listing claim | A seller or manufacturer webpage makes the statement. | Not an independently accepted performance result. |
| Manufacturer document | A linked TDS, SDS or manual has been read. | Each document has a different purpose; an SDS does not prove yield or a fire classification. |
| Report / listing evidence | An identifiable laboratory report, evaluation or certifier listing supports its stated scope. | Only the reviewed claim and configuration, not every use of the product. |
| Identity unresolved | The sold variant and source are not yet fully connected, or relevant fields conflict. | A clarification need, not a finding of fraud or poor performance. |
| Pending / unverified | We have not accepted sufficient traceable evidence for this claim. | Not a negative test result or a different classification. |
Broad research-dimension labels on older records are not claim-level certifications. The dated document checks on product pages identify the narrower source and unresolved questions.
Reviews and advertisements
Before publishing a review-derived quality conclusion, record the platform, exact variant, date window, selection rule, included and excluded counts, duplicates, disclosed incentives and missing installation conditions. A verified-purchase label does not verify installed thickness or long-term performance. No representative review sample or completed failure-rate study is claimed here.
Foam expansion is not a thermal test. A first-day review cannot establish multi-year durability, but lack of follow-up does not prove future failure either. Country of origin, a lower price and frequent advertising are not quality measurements. See how we read owner reviews and expansion demonstrations for the full protocol.
Publication relationships
SprayFoamGuides is our sister publication under common ownership, not an independent third-party endorsement of this site's order. Product-specific material connections must be disclosed alongside any affected recommendation. A final comparative award requires completed evidence and relationship review; a provisional position does not bypass those requirements.
We accept public manufacturer pages, public technical documents and view-only evidence links. Submissions do not guarantee a score change; the research team checks exact-product relevance first.
Corrections policy
When a document, formulation, price snapshot or claim changes the comparison, we update the relevant status and publication date. No brand can pay for eligibility or a ranking position.