Bright Dividenward analyses your earnings patterns and risk tolerance to recommend how supplemental income might be allocated, mitigating volatility without requiring a background in finance.
Gig earnings fluctuate by platform, season, and demand. Deciding how to allocate that income manually means re-evaluating risk each time circumstances change, which is time-consuming and prone to inconsistency.
| Manual Analysis | AI-Optimised Decisions with Bright Dividenward |
|---|---|
| Risk tolerance is estimated once and rarely revisited as income or circumstances change. | Risk profile is recalculated continuously as new earnings and spending data arrive. |
| Decisions rely on whichever data point was most recently noticed, rather than the full picture. | Recommendations draw on the complete, time-stamped history of income volatility. |
| Reviewing options requires dedicated time that gig schedules rarely accommodate. | Analysis runs in the background and surfaces only decisions that require input. |
| Mistakes compound quietly, since there is no consistent benchmark for comparison. | Every recommendation is logged against a stated risk threshold for later review. |
Three components operate together: data ingestion, risk profiling, and execution guidance. Each is designed to be auditable rather than opaque.
The platform processes platform payout records, transaction timing, and relevant market indicators as they arrive. Rather than batching analysis weekly, it flags changes in volatility within the same session, so recommendations reflect current conditions rather than last month's pattern.
Your tolerance for volatility is not fixed. A quieter month, a change in primary platform, or a shift in savings goals all move the score. Bright Dividenward recalculates this probabilistically, weighting recent behaviour more heavily than historical averages, and shows you the direction of the change.
Whether allocating a small surplus after a busy week or a larger sum after a quiet quarter, the underlying logic stays consistent. Position sizing is adjusted proportionally, so the same risk threshold applies whether the amount is modest or substantial.
Transparency matters when a tool informs financial decisions. Each stage below can be inspected within your account activity log.
Link relevant income streams and accounts. Data is read-only at this stage and used solely for analysis.
An initial risk threshold is set from your stated preferences and historical income variability.
The model produces a probabilistic range of outcomes, not a single guaranteed figure.
You approve, decline, or adjust the threshold before anything is actioned on your behalf.
These examples illustrate the type of logic applied, not guaranteed outcomes for any individual account.
A delivery driver earns significantly more during a six-week holiday period than during the rest of the year.
A freelance designer draws income from three platforms with different payout schedules and reliability.
A rideshare driver cuts weekly hours to care for a family member, reducing monthly income by a third.
Bright Dividenward was developed to address a specific gap: most financial analysis tools assume a stable monthly salary. Gig and supplemental earners operate differently, with income that varies by week, platform, and season.
Our approach combines data-driven modelling with a plain-language explanation of every recommendation, so decisions remain yours to make, informed rather than automated without oversight.
Structured answers to common concerns about data handling, risk, and integration.
Only the data required to assess income patterns and risk, typically read-only access to payout records and relevant account balances. No funds move without your explicit approval at each step.
It starts from a short questionnaire and your historical income variability, then adjusts over time based on observed behaviour. You can view and amend the threshold at any point within your account.
No. All outputs are framed probabilistically, reflecting a range of likely outcomes based on available data. Past patterns inform the model but do not determine future results with certainty.
Integration depends on the platforms involved and their available data-sharing permissions. Supported connections are listed during onboarding, and manual data entry remains an option where direct integration is not available.
Data is encrypted both in transit and at rest. Access connections are read-only wherever possible, and you can revoke any linked account at any time from your settings.
Connect one income source and review your first risk-adjusted recommendation. You decide what happens next.
Start AnalysisNo card required to explore the model. Cancel at any point during setup.