Recommendation Intelligence Research™

The State of AI Recommendations in Home & Living

We asked one AI model 20 high-intent home and living shopping questions, 20 times each. This is the fifth and final niche in the series, and it closes the loop on a pattern we have now seen five times.

Atom Foundry · June 2026 · Fifth and final in the Recommendation Intelligence Research™ series
4,000
Recommendations
271
Distinct brands
20
Shopping intents
400
Prompt-runs
How we measured it

Methodology, up front

Identical method to the first four reports, so all five are directly comparable. Everything below is computed from real captured responses. Nothing is estimated or projected.

Category: Home & Living
Model: one model (gpt-4o-mini)
Intents: 20 high-intent prompts
Runs per intent: 20
Prompt-runs total: 400
Recommendations captured: 4,000

Recommendation Share™

A brand's share of all recommendations captured. The whole field sums to 100 percent.

Recommendation Frequency™

In what percent of the 400 prompt-runs the brand appeared at least once.

Recommendation Position™

Average rank in the answer when the brand appeared. Lower is better.

Marketplaces are not brands. Home and living is the most marketplace-driven category we have measured. Retailers and department stores (Amazon, Wayfair, Etsy, Home Depot, Lowe's, Macy's, Nordstrom, and others) were separated out and excluded from the brand-level analysis. Vertical own-brand stores such as IKEA, West Elm, and Crate & Barrel are kept as brands, because they sell their own line through their own store. Cleanups we made and disclose: two brands were hand-corrected after the automatic match picked the wrong store (Crate & Barrel had matched a sub-brand domain, Parachute had not matched at all), and a handful of name variants such as the two spellings of Wsthof were merged. Other limits: one model, one category, one point in time. Each intent ran 20 times. Brands map to the real stores we measure where a match exists; the rest are off-index, never invented.
Who AI recommends

The home & living recommendation leaderboard

Top 20 brands by frequency across all 4,000 recommendations, with retailers excluded. The last column is each brand's real AI Commerce Score™ from our index. The two runaway leaders, IKEA and West Elm, are recommended in nearly half of all prompts while their stores sit only in the Low band.

#Brand Share™ Freq™ Pos™ AI Commerce Score™
1 IKEA 4.8%
 48.3% 5.3 64Low
2 West Elm 4.7%
 47.0% 3.7 60Low
3 Pottery Barn off-index 4.2%
 41.5% 6.0 —Off-index
4 Brooklinen 2.2%
 21.8% 1.5 38AI Invisible
5 Parachute 2.0%
 19.5% 2.8 55Low
6 L.L. Bean 1.7%
 17.3% 6.5 68Low
7 Boll & Branch 1.7%
 17.3% 2.7 51Low
8 Crate & Barrel 1.7%
 17.0% 6.4 50Low
9 CB2 1.7%
 16.8% 5.4 59Low
10 Saatva 1.2%
 11.8% 4.0 51Low
11 Coyuchi 1.1%
 10.8% 5.7 44AI Invisible
12 Tempur-Pedic off-index 1.1%
 10.8% 1.4 —Off-index
13 Casper 1.1%
 10.5% 6.0 73Moderately
14 Purple 1.0%
 10.3% 2.6 56Low
15 Flexispot off-index 1.2%
 9.8% 6.1 —Off-index
16 Ashley Furniture 1.0%
 9.8% 7.5 57Low
17 Autonomous 0.9%
 9.3% 6.2 56Low
18 Herman Miller 0.9%
 8.5% 3.1 28AI Invisible
19 Cuisinart 0.8%
 8.0% 5.0 37AI Invisible
20 Nectar 0.8%
 8.0% 6.6 46AI Invisible
The central finding

Once again, recommendation has no real link to readiness

The correlation between Recommendation Frequency™ and AI Commerce Score™ is r = 0.108 (n = 71), statistically indistinguishable from zero. In a field of 71 single-brand stores, how often AI recommends a home brand tells you essentially nothing about how AI-ready its store is. Each dot is one brand.

02550751000%10%20%30%40%50% IKEA: 48.3% frequency, score 64West Elm: 47.0% frequency, score 60Brooklinen: 21.8% frequency, score 38Parachute: 19.5% frequency, score 55Boll & Branch: 17.3% frequency, score 51L.L. Bean: 17.3% frequency, score 68Crate & Barrel: 17.0% frequency, score 50CB2: 16.8% frequency, score 59Saatva: 11.8% frequency, score 51Coyuchi: 10.8% frequency, score 44Casper: 10.5% frequency, score 73Purple: 10.3% frequency, score 56Ashley Furniture: 9.8% frequency, score 57Autonomous: 9.3% frequency, score 56Herman Miller: 8.5% frequency, score 28Cuisinart: 8.0% frequency, score 37Nectar: 8.0% frequency, score 46Article: 7.3% frequency, score 80Snowe: 6.5% frequency, score 51Humanscale: 5.8% frequency, score 23Steelcase: 5.8% frequency, score 51Anthropologie: 5.8% frequency, score 20Room & Board: 5.5% frequency, score 14Helix: 5.0% frequency, score 57The Sill: 5.0% frequency, score 14Corelle: 5.0% frequency, score 62Uplift Desk: 5.0% frequency, score 77Bath & Body Works: 5.0% frequency, score 14Voluspa: 5.0% frequency, score 46Fully: 5.0% frequency, score 14Secretlab: 5.0% frequency, score 69Global: 5.0% frequency, score 50GreenPan: 5.0% frequency, score 65Wüsthof: 4.8% frequency, score 67Sleep Number: 4.5% frequency, score 48Leesa: 4.0% frequency, score 66Brooklyn Bedding: 4.0% frequency, score 75Bloomscape: 4.0% frequency, score 72Brookstone: 4.0% frequency, score 63Ruggable: 4.0% frequency, score 72Homesick: 3.5% frequency, score 62Eddie Bauer: 2.8% frequency, score 70Joybird: 2.8% frequency, score 65Pendleton: 2.0% frequency, score 45Muji: 1.8% frequency, score 67Cultiver: 1.5% frequency, score 50Ugg: 1.3% frequency, score 14The Citizenry: 1.3% frequency, score 18Rough Linen: 1.3% frequency, score 59Serena & Lily: 1.0% frequency, score 14Buffy: 1.0% frequency, score 58Amerisleep: 0.8% frequency, score 72Nest Bedding: 0.8% frequency, score 69Le Creuset: 0.8% frequency, score 18Molecule: 0.8% frequency, score 26Floyd: 0.5% frequency, score 14Misen: 0.5% frequency, score 64Burrow: 0.5% frequency, score 78Sheex: 0.5% frequency, score 63GhostBed: 0.5% frequency, score 76Sijo: 0.3% frequency, score 58Frette: 0.3% frequency, score 65LAFCO: 0.3% frequency, score 14Nautica: 0.3% frequency, score 53Natori: 0.3% frequency, score 55Magic Linen: 0.3% frequency, score 47Moft: 0.3% frequency, score 60Gibson: 0.3% frequency, score 37Maison Margiela: 0.3% frequency, score 46Lands' End: 0.3% frequency, score 65Arhaus: 0.3% frequency, score 33 Crate & BarrelHerman MillerBrooklinenL.L. BeanParachuteWest ElmArticleCasperIKEACB2 Recommendation Frequency™ (percent of prompts) AI Commerce Score™
Highly (85+)
Moderately (70 to 84)
Low (50 to 69)
AI Invisible (under 50)
The famous home names win on fame. IKEA (48.3 percent of prompts, score 64) and West Elm (47.0 percent, score 60) dominate, and Pottery Barn (41.5 percent) is not even in our index. Meanwhile the best-built stores in the set, Article (80), Burrow (78), Uplift Desk (77), and GhostBed (76), each appear in 7 percent of prompts or fewer.

The average AI Commerce Score™ across all on-index recommended brands is just 50.7, right at the edge of the AI Invisible band. AI is recommending a category of brands whose stores, on average, sit at the threshold of being unreadable to the very agents doing the recommending.

Five niches, one pattern

This was the fifth category, and the answer never changed

Across beauty, supplements, coffee, pets, and now home and living, we have captured more than 20,000 recommendations and measured the same thing every time: the relationship between how often AI recommends a brand and how AI-ready its store is.

CategoryRecommendationsFreq vs Score (r)Relationship
Beauty4,0000.17None
Supplements4,000-0.015None
Coffee4,0000.019None
Pets4,000-0.366Weak negative
Home & Living4,0000.108None
Five categories, more than 20,000 recommendations, and not once is recommendation frequency meaningfully and positively related to store readiness. Three times the link is zero, once it is faintly negative, and here it is zero again. The conclusion is no longer a single finding, it is a replicated result: AI recommends from fame, not from store quality.
What maps to a real store

The most marketplace-driven category yet

Home and living runs through marketplaces more than any category we have measured. Around one in five recommendations went to a retailer or department store such as Amazon, Wayfair, Home Depot, or Macy's, all separated out and kept entirely out of the brand analysis. At the same time, on-index coverage was the highest of any niche, because so many home brands run real stores we have already scored.

~60%
of recommendations map to a single-brand store we actually measure (on-index), the highest of any niche
~21%
go to retailers and marketplaces, the highest share of any niche, separated out from the brand analysis
What it means

Visibility is not recommendation, and recommendation is not readiness

Today AI recommends from memory. It reaches for the household names it saw most during training, IKEA, West Elm, Pottery Barn, even when their stores sit in the Low band or are not in the index at all, and it overlooks the modern direct-to-consumer brands building the most readable, trustworthy stores. That bias is baked into the model, and it is temporary. As AI shopping moves from recalling names to live retrieval and agents that browse, compare, and check out, the advantage shifts to the stores an agent can actually read, trust, and act on.

That is the gap this research exposes and the one the Recommendation Intelligence Framework™ is built to close: AI Readability, AI Understanding, AI Trust, Recommendation Intelligence, and Decision Confidence. The brands coasting on fame today have the most to lose when the model changes. The modern brands building machine-legible stores now are the ones that keep the recommendation when memory stops being enough.

Will AI still recommend you when it has to read your store?
Get your free AI Commerce Score™ in 10 seconds and see exactly where your store stands on the signals that decide AI recommendations.
Get My Free AI Score