loading...
ForMotiv tracks mouse movement, typing rhythm, and pauses to reveal user intent before the form is even submitted.
ENTRY ANGLES
AI analysis of form-fill patterns to detect user hesitation and dropout points in signup/onboarding/checkout flows · Behavioral signal interpretation to identify friction points in digital user interactions · Intent detection from user interaction patterns to generate hypotheses about user behavior
VERTICALS
CAPABILITIES
AI/ML for behavioral signal analysis and pattern recognition, User interaction data collection and interpretation, Product analytics and friction point identification
ForMotiv's premise is straightforward, if unexpected: the way someone fills out a form reveals what they actually intend – and, in some cases, whether they're trying to hide something.
The platform integrates with a website and monitors form completion in real time – tracking mouse movement, typing rhythm, pauses on specific fields, corrections, and hundreds of similar behavioral signals. These observations get transmitted to the site owner alongside the completed form itself.
The AI engine at the core of the platform can recognize roughly 200 behavioral patterns associated with form completion. It has been trained on 500 million completed forms, and each new submission – combined with its eventual manual outcome – feeds continuous self-improvement.
The platform is purpose-built for the insurance market: life, health, home, auto, commercial property, corporate risk, liability, and similar lines. It evaluates each submitted application from two angles.
On the risk side, the platform assigns a risk score based not on the content of the form, but on how the applicant behaved while filling it in. A high risk score can deprioritize the application for review or trigger more intensive verification of the information provided.
On the intent side, the platform flags applications submitted by people who appear to have genuine interest in purchasing a policy – as opposed to just browsing. For these high-intent applicants, it recommends a next action and a channel: send a follow-up email, make a phone call with a specific offer.
The results are tangible. Insurance carriers using the platform report three categories of benefit:
- Lower claims costs through more thorough review of flagged applications.
- Higher revenue from prioritized engagement with high-intent applicants.
- Lower manual review costs by filtering out the highest-risk submissions early.
Adoption is accelerating – the platform's revenue grew fivefold last year. In the current round, the startup raised $1.7M, bringing total funding to $9.7M.
On the surface, this looks like an unusual approach to risk assessment. But it's actually quite intuitive.
In any face-to-face conversation, the majority of information is conveyed non-verbally – facial expression, tone, eye movement, hesitations, the pattern of answers, body language. Even in job interviews, what someone says matters less than how they say it. ForMotiv applies this same principle to digital interaction: not just what someone writes in a form, but how they go about writing it.
Analyzing this at any useful level requires AI. The behavioral signals during form completion are far too subtle and numerous to codify as manual rules. What's needed is pattern recognition across massive datasets, with continuous improvement as that dataset grows – which is exactly what a well-trained ML model does.
The broader idea is compelling: use AI to read the body language of digital interactions and draw inferences about intent and friction.
Insurance applications are one implementation. Loan applications are an obvious next one – arguably with even higher volume and frequency than insurance.
Another natural extension: detecting on which pages of a product or service, and at which specific points, users begin to struggle. Not because they tell you – they almost never do – but because their behavior gives it away: they hesitate, scroll back and forth, move the cursor aimlessly, start typing and then delete. Detecting these micro-signals proactively and optimizing those moments before users quietly give up could be more valuable than any survey or feedback form. Most users don't complain before they churn – they just stop showing up.
E-commerce return prediction is another angle: correlating how someone places an order with the probability they'll return it. A flagged order could trigger a confirmation email a few hours later, reducing costly return logistics.
The direction is fairly clear – using AI to read the "body language" of digital user interactions and generate hypotheses about intent and friction points.
The harder question is where this will be most valuable. Insurance forms are one validated answer. Loan applications are another – higher volume, higher stakes. But the most non-obvious application is probably somewhere closer to the product side: detecting exactly where in a signup flow, onboarding sequence, or checkout users begin to hesitate before they drop off. Most of that friction is invisible until behavioral signals make it legible. ForMotiv's architecture points directly at that problem – it just hasn't been pointed there yet.