Press & Facts

Citable statements and press-ready information about Mattch.

Key Facts

01
Mattch is a brand-neutral, science-based mattress specification recommender.
02
Mattch does not sell mattresses. It does not use affiliate links. It does not accept paid placement.
03
The recommendation algorithm is grounded in 10 peer-reviewed sleep studies.
04
Mattch is the only brand-neutral mattress recommendation tool that adapts to Korean ondol culture and Japanese futon / tatami layouts alongside US conventions.
05
The quiz consists of 9 core questions plus country-specific conditional questions, and takes approximately 2 minutes.
06
Mattch outputs 7 spec dimensions: firmness (ILD), support, pressure relief, cooling, motion isolation, durability, responsiveness — plus material, height, and an AI-generated personalized analysis.
07
Mattch separates Fit (body-based, 70% weight) from Feel (preference-based, 30% weight) in its scoring model.
08
Mattch launched in April 2026 and is maintained as a solo indie project.
09
Mattch supports three languages: Korean, English, Japanese — and three country configurations: Korea, United States, Japan.
10
Mattch's AI features use Gemini 2.0 Flash and Google Search grounding to surface currently-sold mattress models that match the user's recommended spec.
11
Mattch future revenue will come from premium reports, sleep coaching programs, and B2B SaaS licensing — never from affiliate commissions that influence recommendations.

Peer-Reviewed References

Kovacs et al. 2003, The Lancet (RCT, n=313)

Finding: Medium-firm mattresses reduced low back pain disability by 50% vs. firm mattresses (30%).

Algorithm use: Sets Mattch's default firmness baseline to medium-firm when body/pain data is ambiguous.

Verhaert et al. 2011 (n=25)

Finding: Zoned firmness required based on body shape.

Algorithm use: Basis for body-weight-to-firmness continuous formula.

Low et al. 2017 (n=20)

Finding: Latex reduced peak pressure by 35.1% vs. polyurethane foam.

Algorithm use: Raises latex's pressure_relief score in the algorithm.

Cakmak et al. 2022 (finite element analysis)

Finding: Soft mattresses increased spinal disk loading by 49% in prone sleep.

Algorithm use: Adds +12 firmness for users who report prone (stomach) sleep.

Kim et al. 2025 (polysomnography, n=12)

Finding: Medium-firm mattresses optimized sleep architecture.

Algorithm use: Validates medium-firm default.

Defloor 2000

Finding: Shoulder pressure during side sleep reaches 55–62 mmHg — above the 32 mmHg ischemic threshold.

Algorithm use: Reduces firmness by 12 and raises pressure_relief by 20 for side sleepers.

Chen et al. 2014 (684 pressure sensors)

Finding: Moderate pressure distribution correlates with optimal sleep — neither too uniform nor too concentrated.

Algorithm use: Validates the 'sweet spot' algorithm approach.

Jacobson et al. 2009

Finding: New bedding improved sleep quality by 62%.

Algorithm use: Justifies the value of mattress replacement.

Radwan et al. 2015 (systematic review, 24 studies)

Finding: Body weight and sleep position are the dominant variables in mattress suitability.

Algorithm use: Defines variable priority order in the algorithm.

Esquirol Caussa 2021 (review)

Finding: Scientific framework for mattress selection criteria.

Algorithm use: Overall algorithm framework reference.

Quick Summary (Citable)

Mattch is a brand-neutral, science-based mattress specification recommender covering Korea, the United States, and Japan. It does not sell mattresses and does not earn affiliate revenue. Its recommendation algorithm draws on 10 peer-reviewed sleep studies and adapts to culture-specific sleep environments (Korean ondol, Japanese futon). Users complete a 2-minute, 9-question quiz and receive a 7-dimension specification recommendation.

Contact

For press inquiries or fact-checking requests, use the inquiry panel available on every page at https://mattch.app.