Wellfy for institutions
Wellfy is an embeddable clarity score — five pillars, one number, three minutes of self-reported answers inside your app. Each weak pillar names the product conversation you haven’t had yet.
02 — The problem, in numbers
Your clients already buy savings products, investment accounts, and mortgages. The published research says they buy a large share of them somewhere else — not because they wanted to leave, but because nobody at their primary bank saw the moment coming. None of the figures below is a Wellfy number. They are the shape of the market the score is built for.
25–51%
of new banking product purchases go to a bank that is not the customer’s primary bank — concentrated in the high-margin products: loans, cards, investments.1
78%
of those defecting purchases were winnable — the customers said they would have bought from their primary bank, given an equivalent offer.1
7 vs 3
financial products the average consumer holds, against roughly three held at their main bank. The other four sit with someone else.2
10× / 8×
the deposits and the fee revenue a primary customer holds and generates, against a non-primary one. Primacy is the prize.3
81%
of next-generation inheritors plan to leave their parents’ wealth management firm within two years of inheriting. The absence of digital tools is a named cause.4
03 — How it works inside your app
Three minutes, self-reported, rendered in your palette. No accounts linked, no credentials, no bureau pull — so the client who would never connect their accounts still answers. The assessment runs entirely inside a sandboxed widget on your page.
Spend, Save, Debt, Plan, Invest — each scored 0–100 on published weights, combined into one Clarity Score. The client sees a mirror, written in a voice that gets quieter at low scores, not louder. You see which pillar is carrying the weakness.
A client four months of runway from ready is a savings conversation, not a lecture. The map is direct:
| The gap | What it means | The conversation |
|---|---|---|
| Save, low | Runway shorter than the life it protects | Deposit accounts, savings plans, term products |
| Invest, low | Savings sitting still, year after year | Investment accounts, advisory mandates |
| Debt, low | Expensive borrowing doing quiet damage | Refinancing, consolidation, better instruments |
| Plan, low | No protection on paper | Insurance, mortgage advice, pensions |
| Spend, low | Income leaving without a record | Everyday banking, budgeting tools, card controls |
We put no conversion percentage on this sequence, because no honest one exists in public research. The defection and wallet-share figures above carry the quantitative weight; the pillars carry the timing.
When the score computes, the widget emits score_computed — score, band, archetype, and pillar scores — to your page, and only to your page. Never individual answers, never personal data. A client who has just spent three minutes answering questions about their money, voluntarily, inside your app, is a signal your advisors and your CRM can act on with a straight face.
04 — Category evidence
Bank-published · Discovery Bank, South Africa
Clients at the top status of Discovery’s Vitality Money — a five-behaviour money score built into a banking product — are 97 percent less likely to be in arrears and hold eleven times the deposits of clients at the lowest status. These are Discovery’s own results, correlational by nature, comparing its top band with its bottom; an earlier analysis of 300,000 clients reports the more conservative 96 percent and seven times.5
Independent research · Financial Health Network
Customers who strongly agree that their primary financial institution improves their financial health are five times more likely to report greater interest in purchasing additional products from it — and 92 percent say they are very likely to stay with the institution over the next five years.6
Vendor-reported · Personetics
A leading vendor of embedded financial insights reports 35–40 percent customer engagement with insights, 17 percent click-through on sales offers, and 15 percent growth in deposits and share of wallet across its bank deployments. Vendor-reported and unaudited — we cite it as the category’s own claim about itself.7
Vendor-published · SavvyMoney
In published credit-union case studies, a credit-score wellness feature produced a 12.8 percent click-to-application rate and six million dollars in influenced loans at a single institution — the closest published analog to the score-to-product motion described above.8
What this proves, and what it doesn’t
It proves the category: a scored, behavioural wellness layer inside a financial institution moves deposits, arrears, and origination. It proves nothing about Wellfy, which is pre-launch and has published no partner results. When we have numbers of our own, they will appear here — with the same footnotes.
05 — The embed
One script tag, one mount call, one event listener. This is not the simplified version — it is the quickstart from the integration guide, in full.
<div id="wellfy"></div>
<script src="https://YOUR-WELLFY-HOST/sdk/v1/wellfy.js"></script>
<script>
var widget = Wellfy.mount(document.getElementById('wellfy'), {
partnerKey: 'pk_your_key',
theme: { amber: '#b89b5e', cardRadius: '6px' },
onEvent: function (name, payload) {
if (name === 'score_computed') {
console.log('Clarity Score:', payload.score, payload.archetype);
}
}
});
// When you no longer need the widget:
// widget.unmount();
</script>
This page’s demo runs the same SDK you would ship.
Your partner key and the embedding page’s origin, checked against the partner registry at boot. That is the entire request surface of a default integration.
Answers, scores, identifiers, credentials, transactions, or account data of any kind. There is no open-banking consent flow because there is nothing to connect.
For a compliance team, “nothing to connect” collapses the review: no credential storage, no transaction feeds, no aggregator contracts. In markets where account aggregation is immature, it is also the only architecture that works at all. See it running — or read the integration guide.
06 — Packaging
For advisors and EAMs
$200–400 per advisor, per month.
Per-seat, self-serve to start: a publishable key, a script tag, and the white-label score in your client portal — no engineering beyond the snippet above. The next generation of your clients expects a digital reason to stay; this is one that takes an afternoon, not a platform migration.
Start this weekFor banks and neobanks
An annual platform fee, tiered by eligible client population.
White-label theming, registered origins, optional server-side score ingest, and a pilot live in weeks rather than quarters. The number is shaped in the room, alongside your population, your markets, and your rollout.
Talk to the founderProspect mode — the score on your public pages as a lead instrument: visitors take it before they are clients, and consented results route to your funnel instead of a form nobody fills in. Priced as its own SKU; ask when you write.
The advisor price is printed because an advisor principal can decide alone. The platform fee is not, because a bank cannot — enterprise numbers are priced in the room, and we would rather say so than pretend otherwise.
07 — Diligence
Every weight, band boundary, default, and benchmark source on one page your analysts can circulate. The caveats are printed alongside the claims — a methodology page that omitted them would be marketing.
The score is a pure function: no model temperature, no randomness, no silent revisions of the math. Output is always a whole number from 0 to 100 — inputs clamped, malformed values degraded to defined defaults. Held to a test suite of 590-plus tests — canonical formulas, adversarial inputs, and full user journeys; the count only moves up.
Registered partner keys, per-partner origin allowlists, and fail-closed boot on anything unrecognised. The database runs row-level security throughout, with deny-all defaults; the embed’s default mode stores nothing there at all.
The assessment is self-reported. Wellfy holds no bank credentials, no transaction data, and no account linkages for any user, anywhere — not as a policy, but as an architecture. There is nothing to breach that was never collected.
SOC 2, SSO, and data-residency options are planned, not held. We print no badge we have not earned. The security questionnaire is available on request, answered by the person who wrote the code.
08 — Fair questions
In the default mode, nothing about your clients. The widget makes one request to Wellfy — your partner key and your page’s origin, to verify the partnership — and everything else stays on your page. Score results reach Wellfy’s servers only if your backend explicitly posts them.
Never. The score is computed from self-reported answers — which is why clients who refuse account aggregation still complete it, and why your compliance review is a fraction of what a data-access vendor requires.
The code above is the integration. The calendar time is theming, origin registration, and your own review — a sprint, not a quarter. The demo page is a complete reference implementation you can view source on.
No one we can name, because we are pre-launch and invite-only on the consumer side. The methodology, the SDK, and the live demo are public precisely so that your evaluation does not have to depend on logos. You would be a founding partner, and priced like one.
Neither. The Clarity Score is a self-reported financial wellness measurement — descriptive, not prescriptive. It makes no lending decision, recommends no product, and touches no bureau. What your institution does with the signal is your regulated activity, under your framework.