Resonance · the message validation platform
Every message is judged the second it lands. Read the verdict first.
Resonance scans your communication through the audiences who will actually read it and returns three readings — Exposure, Clearance and Reception — each finding pinned to the exact phrase, explained by a mechanism, cited to a source, rated for confidence, and paired with a specific rewrite.
“A fresh start — leave the old ways behind you.”
- How it read
- For first-generation audiences, “leave the old ways behind” reads as a demand to discard heritage. The campaign becomes the story.
- The mechanism
- The framing conflicts with bicultural identity integration: for much of this audience belonging is additive, not substitutive.
- The change
- “A fresh start — bring what matters with you.” Keeps the promise, removes the demand.
Benet-Martínez & Haritatos (2005) · confidence: high · reviewed lens
The problem it solves
A good message is not a good message to everyone.
Every brief carries the same silent assumption: that a message which works, works. It is the assumption behind the sign-off, the media plan and the launch date.
The average protects that assumption. Aggregate testing returns one number. A 68 looks acceptable, so the work goes out. But 68 is the mean of an 88 and a 48: one audience you moved, and one you alienated. Nobody experiences an average.
By the time a brand-lift study or a focus group tells you otherwise, the money is committed and the message is public. Exposure is not something you can un-spend.
- What aggregate testing returns
- 68 — looks fine. Publish it?
- What resolution returns
- Lens A: 88 · Lens B: 84 · Lens C: 41.
Two audiences moved. One lost. The number never showed you which.
aggregation hides · resolution reveals
Three readings, one scan
Different teams need different evidence from the same message.
Every scan returns all three. Nobody has to decide in advance which question matters, and legal, compliance and marketing work from one set of evidence instead of three separate reviews.
Reading 01
Exposure
“What will this cost us?”
The element that becomes the headline, found while it can still be changed. You get the reason it misfires, the audience it misfires for, and the source behind the call.
Corporate affairs · Communications · Legal · Brand safety
Reading 02
Clearance
“Can we sign this off?”
The check you run on every message, every time. Versioned, re-runnable and defensible upward, so “was this reviewed?” has a record for an answer.
Compliance · Brand governance · Risk · Regulatory affairs
Reading 03
Reception
“Will this actually work?”
Your message broken out per audience, so you see which one you are losing, why, and the change that lifts the weakest read without breaking the strongest.
Marketing · Campaign owners · Creative leads · Agencies
The loop
A scan is a report.
A re-scan is a checkpoint.
One-off analysis changes a decision. A loop changes how the work gets made.
Add your message
Upload the draft. Copy, script, letter, policy, subject line.
Choose your lenses
Select the audiences the message has to reach. Each lens is one audience, resolved.
Read the findings
Every finding pinned to the phrase that caused it, with mechanism, source, confidence and rewrite.
Change it and run it again
The re-scan shows what moved, and records that the message was reviewed before it went out.
The record remains after the campaign ends. Every scan is versioned: what was checked, against which criteria, what changed between drafts, and what cleared. When someone senior asks whether this was reviewed, you can produce the document.
What sits underneath
Lenses are the unit of resolution.
A lens is one audience, described by what actually governs how it reads: whether identity is held individually or collectively, whether people move toward gain or away from risk, how much effort the wording demands, how much trust the sender starts with, and how strongly being told what to do provokes resistance. Not a demographic label. Not a persona invented in a workshop.
Criteria are community-led. Weights are expert-led. The evidence in a lens comes from the communities it describes and the researchers who work with them; how heavily each factor counts is a methodological judgement we own and publish.
Lenses are versioned. When the evidence behind one changes, the lens changes, and every scan carries the version it ran against.
Standard lenses
Built and reviewed by Mosaica from the Corpus, covering the audiences most organisations need to reach.
Curated lenses
Tuned for a sector such as healthcare, government or financial services, so findings are reported in the vocabulary that sector actually uses.
Proprietary lenses
Built from your own audience evidence or a partner’s, and available only to you. Your data stays yours and never trains public models.
Frequently asked questions.
Is this just a general AI model with a prompt?
No. Frontier AI does the scanning and structures the evidence — that part is not the product. What Mosaica governs is the corpus it reads from, the criteria it applies, the confidence thresholds, the red lines, and the expert review behind the sources. Anyone can generate an opinion about an audience. Almost nobody can show you the source, the mechanism and the confidence behind it.
How is this different from stereotyping?
We never say a group believes something. We identify a mechanism — an interdependent self-construal, a prevention-oriented regulatory focus, a low institutional-trust gradient — which is more prevalent in a cohort, is documented in the literature, and varies by generation, acculturation and context. A mechanism is testable, citable and defensible. A stereotype is none of those. Every finding names its mechanism and cites its source, so you can check the reasoning rather than take the verdict.
Does it replace research, or a focus group?
No, and it is not intended to. A focus group tells you what eight people said in a room. Resonance tells you which element of your message conflicts with which documented mechanism, for which audience, with the source attached — in the time it takes to read a draft, at the point where changing it is still free.
Is a machine making the final call?
The scan produces findings; people make decisions. Every finding carries a confidence rating and a citation precisely so a person can weigh it. Sources are expert-reviewed and community-consented, red lines are held as rules rather than model judgements, and there is a correction pathway when we get something wrong.
What happens to our messages and our data?
Your content is yours. It is not used to train public models, and proprietary lenses built from your evidence stay exclusive to you. The full data-handling and security posture is on the Trust and security page, including documentation available for ICT and procurement review.
See it run on your message.
Bring one real message going live this month. Twenty minutes, and you keep the findings.