Reduce defaults without shrinking the book.

No. Zoral fOS can read from the systems you run today, use your existing models alongside its own, and return its proposals to them. A common starting point is to track one portfolio against plan and add early warning, then decide what else to move. You decide what you replace, and when.

Cutting approvals is the bluntest tool. Zoral works on two fronts. On the existing book, accounts are re-scored, and cut-offs, limits, pricing and collections are recalibrated, which frees funding tied up in weak lending. On new lending, robo-underwriting or assisted underwriting decides applications consistently against your policy, so new loans are better underwritten. As the new book grows, the default rate of the whole portfolio can fall while total volume rises. Results are tracked against plan, so you see the effect on volumes, margin, losses and customer outcomes. Your credit policy sets the trade-offs.

They are built to be challenged. Each model is documented, tested on data it was not trained on, versioned, monitored for drift and stability, and can run in shadow or as a challenger before it goes live. Scores come with reasons your teams can explain to a customer or a supervisor. Your own risk function validates and approves them.

The purpose of early warning is to offer help sooner. Contact frequency, channel and timing follow your policy and the customer’s preferences. Forbearance options are offered where your policy allows, and customers who may be vulnerable are routed to trained staff. You can compare outcomes across customer groups and show how customers in difficulty were treated. Your obligations remain yours, and Zoral gives you the controls and evidence to meet them.

They could, which is why both pass the same affordability and suitability tests as a new application. Customers showing signs of financial difficulty or persistent debt can be excluded, marketing preferences are respected, and every offer is recorded with the reason it was made. You decide which offers are made, and to whom.

Your people do, under the authorities you already have. A proposed change is first run against past accounts to show its likely effect, then signed off by a second person or by committee, as your delegated authority requires. It can start on a test group before reaching the whole portfolio, and it can be rolled back with its record intact.

Yes. Each product’s plan covers revenue, acquisition and data costs, operating cost, losses, recoveries and capital used. Actual results are compared with it by product, channel, partner or region. Where some figures, such as operating costs, arrive only monthly, agreed baselines can be used until they do. Your finance and risk teams agree the plan. The platform keeps score.

Agents, configured to your policies and workflows, help analysts prepare scenarios and committee commentary, and help collections staff prepare for customer conversations. They act only through the permissions, workflows and approvals you set, and what they do is recorded. Agents do not make credit decisions. Credit decisions follow policy and models your bank approves, and accountability stays with your bank. You choose the language models behind them, connected to Zoral fOS through a standard interface (MCP). Client data, knowledge bases and conversation records stay under your governance, within the environments and agreements you control. Separately, we use AI for Zoral to help our specialists and your analysts design and test strategies more quickly.

Usually not. Zoral connects to your core banking, origination, card and collections systems, payment systems, bureau files and data lake through documented interfaces, by API, messaging, events or file transfer. Data is held in a banking data model with lineage, so each figure can be traced to its source. Start with the data you have. We show where better data would strengthen the models.

Zoral runs on AWS or Azure in your chosen region, or in your own data centre, and is designed for high availability. If you use it, Zoral Document Fabric, which reads your documents, runs in your own AWS or Azure account, or your private cloud, in your region and under your control. Our controls are independently audited under SOC 2, with details on the Zoral Trust Portal. The platform is designed to support GDPR and UK GDPR obligations. Access to customer and model data follows the roles you set. Architecture documentation is available under a confidentiality agreement.

Zoral can run and support the platform for you, your own technology team can run it, or we share the work and hand over as your team grows. Your people run the banking operations, including credit decisions and collections. Our data scientists can build and maintain models with your team, and hand them over as your team takes them on.

That depends on the portfolio, the data available and how often your committees meet. Start with one portfolio: agree the plan, back-test the models and strategies on its history, then go live on a test group. We give a realistic plan after scoping, not before.

Zoral is an independent specialist, founded in 2004, with over 100 projects and 350+ years of combined experience, and our platform is our own. We have worked with clients regulated by the FCA in the UK, the CFPB in the US and the central banks of Malaysia and Thailand, and we have a long-standing relationship with KPMG. Your data, product definitions and audit history remain yours and can be exported in standard formats. References are available in confidence.