Ivernholl analyses market data, transaction history and income patterns in real time, then adjusts its recommendations to match how much variance you can actually absorb — not a generic, one-time risk score.
A continuous view of exposure, variance tracking and recommendation confidence, recalculated as new data arrives rather than on a fixed daily schedule.
The system is not a static calculator. It ingests data continuously and recalculates recommendations as conditions change, keeping outputs aligned with current, not historical, risk exposure.
Market feeds, transaction histories and income patterns are processed continuously rather than batched into a fixed schedule.
Update cycle: ContinuousRecommendations are weighted against your observed tolerance for drawdown, derived from behaviour rather than a static questionnaire.
Recalibration: OngoingOutputs adjust as your financial situation changes — shifting income, new liabilities, or altered short-term goals.
Inputs: Multi-sourceAnalysis runs close to the data source, so recommendations reflect present conditions rather than a prior closing figure.
Response: Sub-secondIvernholl was designed around a simple constraint: most individual investors and gig-economy professionals do not have the time to monitor markets continuously, nor the staff to manage a varied risk mandate across departments.
The platform takes over that monitoring function. It applies the same class of predictive modelling used in institutional risk desks, scaled down to operate on an individual account or a single corporate portfolio, and adjusted automatically as circumstances change.
Most platforms assign a risk category once, at onboarding, and leave it unchanged. Ivernholl treats risk tolerance as a variable that moves with behaviour, income stability and observed reactions to volatility.
The practical result: recommendations narrow in on a risk level you can sustain, rather than one you selected once and may have misjudged.
The same underlying model supports two distinct situations: individuals allocating variable, part-time capital, and corporate teams managing several risk mandates at once.
A gig-economy professional with irregular weekly earnings allocates a portion of available capital through Ivernholl rather than managing positions manually between jobs. Contribution size varies week to week, and the model adjusts position sizing to match available capital rather than assuming a fixed deposit schedule.
A corporate finance team oversees several internal portfolios, each with a different approved risk ceiling. Instead of maintaining separate spreadsheets per mandate, Ivernholl applies mandate-specific variance limits within a single monitored view, flagging any portfolio that drifts outside its approved range.
The workflow below describes how raw data becomes a usable recommendation. Each stage is logged, so any output can be traced back to the inputs that produced it.
Account and transaction data is encrypted in transit and at rest. Access to raw data is logged and restricted to the processes that require it for model operation.
Recommendations refresh as new data is ingested, rather than on a fixed daily or weekly cycle, so outputs reflect current rather than stale conditions.
The model combines an initial, questionnaire-based baseline with ongoing behavioural data — how recommendations are accepted, adjusted or declined over time. The resulting score is recalculated continuously rather than fixed at onboarding.
Yes. The model does not assume a fixed deposit schedule. Position sizing is recalculated against whatever capital is currently available, so variable income does not require manual recalibration each time it changes.
Typically, transaction history and account-level data from connected sources, along with any market data relevant to your holdings. No data beyond what is required for risk scoring is requested.
Yes. Every recommendation can be adjusted or declined manually. That response is itself treated as a data point, feeding back into the variance model for future recalibration.
Data is encrypted in transit and at rest, with access restricted and logged at the process level. Security practices are reviewed as part of ongoing platform maintenance.
Yes. Each mandate can be scored and monitored independently within a single account, with variance limits applied per mandate rather than across the portfolio as a whole.
Connect a data source and the model begins learning from your actual transaction behaviour, not a static questionnaire result.
Deploy AIAccount setup typically takes a few minutes; the model begins calibrating from your first connected data source.