# Find the information behind an answer When you ask how much is outstanding or what changed on a job, a convincing sentence is not enough. You need to know which records were included, what was calculated, and whether a newer update exists. Areinum uses a graph to find relevant company records and tools for lookups and calculations. The assistant brings the resulting information into the conversation so you can inspect the answer and decide what to do. Canonical: https://areinum.com/product/data-checks/ Language: en Updated: 2026-10-31 ## Give the question a clear scope A question such as "Which invoices are overdue?" needs a date and a definition of what counts as overdue. It also needs the relevant invoice and payment records to be available. State the customer or period when you want a narrower answer. If a payment is present in one record but its link to an invoice is unclear, keep that uncertainty visible. A complete-sounding total should not conceal an unresolved match. ## What the graph and tools do A persistent graph indexes metadata and relationships from company data. It can help locate records, group or count them, and identify selected source text to read when the question requires more detail. Business tools and parts of the platform are written in Rust. The point for an owner is that code can handle defined operations, including calculations and file work, while the assistant explains the result. Rust by itself does not prove a particular response time or accuracy rate. ## An example: which invoice needs attention? Suppose a fictional customer has three invoices: one paid, one due next week, and one past its stated due date. An illustrative answer would list the overdue invoice, its amount and due date, and the source used to determine its status. If the latest payment information is missing, the useful result says the status needs confirmation before a reminder is sent. The example shows the kind of question to ask; it does not claim a measured performance result or a real customer outcome. ## Review the answer in four moves Look at the scope and the source, not just the figure. Dates, currencies, duplicate records, and missing payments can change an otherwise correct calculation. When the answer leads to a message or a change in a system, review that proposed action separately from the calculation. Record what was actually completed after the decision. 1. Check the customer, date range, and definition used. 2. Open the source behind a consequential figure. 3. Resolve missing, duplicate, or conflicting information. 4. Approve a separate next action only after checking it. ## Questions you may have ### Does Rust guarantee that answers are instant? No. Rust is part of the implementation, not a response-time guarantee. A speed claim would need a measurement for the task and conditions in question. ### Will every calculation be correct? Defined operations can be carried out by tools, but the result still depends on the inputs and the operation chosen. Check the source values, date range, units, and unresolved records before using a consequential figure. ### Can I see where an answer came from? Demo answers can include source information and a data-up-to marker. Use those details to inspect the relevant records and decide whether a newer update is needed. ### Does no result mean the event never happened? No. It may mean that the available records do not contain a match or that the search needs a different scope. Check another appropriate source before treating absence as a fact. ## Related reading - [Find the details behind a business question](https://areinum.com/product/company-knowledge/) - [Review the next step before it affects someone else](https://areinum.com/product/decisions-and-control/) - [How Areinum finds and checks information behind an answer](https://areinum.com/guides/how-areinum-finds-and-checks-information/) Explore the demo: https://platform.areinum.com/ The demo uses sample business data and is in English. Access may need to be arranged.