Zarađaština analyzes your income, expenses and relevant market data in real time and automatically suggests when it is reasonable to reserve funds and when to reinvest them.
No obligation. The data remains under your control.
One month brings more projects than you can handle, and the next is left empty. That pattern isn't the result of poor planning — it's the result of a market that changes faster than one person can keep up. Zarađaština was created to fill that void with analytics, not guesswork.
Service prices, sector demand and competition change weekly. It is difficult to distinguish a temporary drop in demand from a permanent trend without constant data monitoring.
Without a stable salary, the decision about how much to save and how much to invest carries a greater burden. A misjudgment of risk is felt more quickly when the income varies from month to month.
Time spent on market analysis is time not billed to the client. Most part-time professionals simply don't have the capacity for in-depth financial analysis with a full-time job.
Instead of general advice, Zarađaština processes your real data — revenues, costs, seasonality of projects — and combines it with publicly available market indicators. The result is concrete, scalable recommendations that adapt to your situation, rather than a generic user average.
Based on this, the model suggests allocating part of the surplus to a low-risk reserve before the next seasonal downturn, with an estimate of how long that downturn could last, according to historical patterns.
Zarađaština was developed as a financial management tool for those whose income depends on projects and not on a monthly salary. Instead of offering universal advice, we train the model on aggregated, anonymized data from a community of users who voluntarily participate in checking the results.
The approach is deliberately conservative: the priority is risk reduction and transparency of the methodology, not promises of rapid growth. Each recommendation is based on data that is available for verification and not on assumptions that the user cannot see.
The logic of the model does not remain closed in the system. Each stage, from the collection of data to the public announcement of the results, goes through a process that can be monitored and verified.
Data on revenues, costs and market trends are collected from voluntary user inputs and publicly available sources.
Before any analysis, personal identifiers are removed. The model works with forms, not with the personal data of an individual.
Some users voluntarily compare model recommendations with actual outcomes and report deviations that are used for corrections.
Aggregated, anonymized results are published in a public record so that anyone can monitor the actual accuracy of the model over time.
The record is updated at the conclusion of each cycle and remains publicly available, including cases where the model recommendation has been corrected following community feedback.
The model monitors demand trends in your service category and warns when market prices are on the rise or fall. Instead of determining the price by intuition, you make a decision based on the current state of supply and demand.
Users who adjust prices according to the recommendations report a steadier influx of new inquiries during periods of lower seasonal demand, according to public record data.
When the model detects a period of stable surplus income, it suggests allocating part of that surplus to low-risk instruments, while estimating how long the reserve should cover a possible period without projects.
The goal is not to maximize returns, but to reduce the need for emergency borrowing or asset sales during periods of low demand.
The data you enter is used to generate your personal recommendations and, with your consent, for anonymized aggregation that improves the model for the entire community. Personal identifiers are removed prior to any aggregate analysis, and participation in the aggregation is not a requirement to use the core recommendations.
The model distinguishes short-term oscillations from structural changes by comparing new data with historical patterns spanning multiple periods. In case of unexpected deviations, the recommendations become temporarily more conservative until enough data are collected for a more reliable assessment.
That. Recommendations are adjusted to the amount you actually have available, and the model prioritizes building a reserve before considering investment. There is no minimum amount required to use revenue and expense analytics.
Each month of work creates data that, analyzed in time, can help make the next financial decision. Join the community that turns that data into validated, publicly available insights.