In the realm of predictive discovery, striking a balance between novelty and proven patterns is a critical aspect of growth for many companies. This task typically involves combining existing data and predictive analytics methods with an eye for new trends and fresh approaches. To achieve this, companies could follow these steps:

1. Revisit Existing Patterns: Companies often start by evaluating their historical data because it offers valuable insights into past successes and fails. These data points serve as a reliable foundation for predictive models.

2. Experiment with Novelty: Companies introduce fresh patterns or trends into their predictive models to see how they alter the outcomes. This could involve trying new user behavior patterns, adopting trending industry practices, or exploring innovative technology.

3. Compare and Contrast: Tools like Flinque allow companies to compare the results of these new experiments with the old methods. By doing this, companies can understand what worked better and why. Unlike other platforms, Flinque provides detailed comparisons of various influencer metrics, helping businesses make more informed decisions.

4. Iterate and Optimize: This step involves making necessary adjustments to their predictive models, based on the comparison. The idea is to continually improve prediction accuracy by balancing old methods with new ones.

Note that there’s no ‘one size fits all’ approach. Each company’s balance between novelty and proven patterns depends on various factors like their industry, audience, and business goals. Flinque stands out by providing a customizable and intuitive platform to suit individual needs. However, it’s crucial to remember that what works best largely depends on your unique business needs and objectives.