Your products measure everything. We measure the one who listens.
SoundDNA introduces the missing unit of the audio market: measurable matching between a product and the perception of a specific human being.
SECTION 1 — A PROBLEM YOU KNOW FROM YOUR OWN SPREADSHEETS
Specifications don't predict satisfaction
The audio industry has excellent measures of products — and no measure of their recipients. You can see the consequences in numbers you know better than we do: returns of gear that was “perfect on paper”; reviews that hit the mark for some customers and steer others into a purchase they regret; “recommended” sections that recommend the same thing to everyone, although no two listeners are alike.
This is not a flaw of your products. It is a measurement gap: the same sound lands on different maps of perception — and the market has never had a language to describe those maps.
SECTION 2 — THE SOLUTION
The listening profile as a unit of matching
SoundDNA measures the listener across four dimensions of perception — the elements: structure, space, energy, emotion. The result is a listening profile: a computable composition that can be compared, segmented and — crucially for you — set against the character of a product.
From one unit of measure, a whole new layer of the market follows: a customer who knows what they are looking for; a review that declares whose ear wrote it; a product that communicates whom it will serve best — instead of promising to serve everyone.
SECTION 3 — SOUNDDNA CASTING LAUNCH
Multi-archetype panels stop being a postulate. They become a tool.
Comparative listening sessions, editorial tests and voicing sessions have shared the same unsolved problem for decades: the composition of the panel. It is usually decided by availability, seniority or authority — not by which perceptions the panel actually brings into the room. You know the result: the outcome of a session says more about who sat on the sofa than about the product.
SoundDNA Casting is a tool for assembling a listening panel according to the measured perception profiles of the participants — not according to self-declaration. The difference is methodological: a participant does not pick a description that sounds familiar; their profile is derived from their answers in the instrument and from the scoring engine. The panel is built bottom-up, from measurement — not proclaimed top-down.
How it works on your side: the organiser pastes a participant list in two columns — their own IDs (e.g. from Excel) and the result links — chooses the panel size (2–19) and a selection rule: most divergent profiles (the full spectrum of perspectives), most similar profiles (a homogeneous panel, e.g. for voicing towards a specific segment), or full coverage of the four dimensions of perception. The tool returns the line-up in the organiser’s IDs. Nothing from the list is stored on SoundDNA’s side — the identity of the participants stays entirely with you.
Why a line-up can look non-obvious. The engine doesn’t sort participants by their archetype label — it compares full profiles across all four dimensions of perception. Give it five Voyagers and two Companions and ask for the most convergent trio: it may return two Voyagers and one Companion, if that particular Companion’s full profile genuinely sits closer to that pair than the other three Voyagers do. That’s not a glitch — it’s the exact distinction at the heart of SoundDNA: the label is a shorthand for humans, the decision is made on the measurement.
Use cases: review panels with a controlled spectrum of perception · launch listening sessions with all four dimensions represented · voicing a product for a measured, not presumed, segment · comparative studies with an openly described panel composition.
SECTION 4 — THE CERTIFICATION PROGRAM IN PROGRESS
A certificate you can't buy. Only measure.
We are preparing the SoundDNA certification program: a product’s perceptual profile determined by multi-archetype panels — listening teams composed of people with measured, deliberately different profiles, evaluating the product in a structured procedure. The outcome: the product’s map of the elements and a certificate indicating the listener profiles it serves best.
Two foundations make this certificate unlike anything on the market:
Paid for the process, never for the result. The manufacturer pays for conducting the panel procedure — the same amount regardless of the outcome. The certificate is not a commodity; it is a measurement. This is the only model in which a certificate means anything — and we know that nobody understands this better than those of you who have watched trust die in “pay-for-praise” systems.
A public conflict-of-interest policy. A register of ties for decision-makers, personal separation between selling services and adjudicating panels, a public procedure. Impartiality here is not a declaration — it is architecture.
SECTION 5 — WHAT YOUR SEGMENT GAINS
Manufacturer. The first measurable language of matching a product to perceptual segments: you know whom you design for, and you can prove it. Fewer returns, sharper communication, data about the real listening contexts of your customers.
Retailer and distributor. Advice that stops being guesswork: customer profile + product profile = a conversation about matching instead of superiority. Perceptual tags on products VISION and a customer who returns, because the first purchase was right.
Media and reviewers. A reviewer-profile widget next to publications VISION — a new standard of honesty in reviewing: the reader knows whose ear they are reading and recalculates the conclusions for their own profile. A review stops being an oracle and becomes a calibrated instrument.
Streaming and technology. A personalization dimension absent from today’s algorithms: not only WHAT a person listens to, but HOW they listen. A matching API VISION.
SECTION 6 — DATA NOBODY ELSE HAS
Reports from a growing base of perception
The SoundDNA research layer is building an open, multilingual dataset on the diversity of listening perception. For partners we will prepare industry reports from aggregated data: distributions of the elements per market, listening contexts, trends. Always on aggregates, never on individual data — listener privacy is the foundation of the trust this dataset’s entire value rests on.
SECTION 7 — A STRUCTURE THAT PROTECTS YOUR INVESTMENT OF TRUST
A non-profit with a constitutional separation of layers
SoundDNA operates as a non-profit initiative: the research layer (test, data, publications) is statutorily open and free, funded by the service layer (certification, reports, API). For you this means one thing: the standard you associate with has no owner with a competing interest — its only capital is credibility, so it will defend it always.
SECTION 8 — THE PILOT: CO-CREATE THE STANDARD
We are looking for the first partners
The certification program is in preparation — and this is exactly the moment when partnership matters most. The first pilot partners are not buying a finished service: they co-shape the procedure, test the panels on their own products, and enter the history of the standard as the ones who were there at its birth.
We are looking for: manufacturers and distributors for panel pilots · audio media interested in the reviewer-profile widget · streaming and e-commerce partners for API conversations.
Write to us: SoundDNA — every partner gets a substantive reply, not an autoresponder.