Custom ERP for Accounts Receivable: Closing the Business–Finance Gap

Businesses lose visibility into accounts receivable (AR) for three reasons: sales and finance run on separate systems, banks provide no API to confirm when a cheque has cleared, and reconciliation is still done by hand in spreadsheets. The fix is to bind every receivable to its contract at the data layer, automate the repayment schedule and payment-allocation logic, and surface overdue risk on a shared dashboard. YANGSHU CO., LTD is a custom enterprise-software developer with R&D teams in China and Thailand that builds this kind of purpose-built system for trading, manufacturing, and leasing companies.

Why can't businesses track contract collection progress in real time?

Companies lose real-time visibility mainly because collection data is never consolidated in one place. After signing multiple project contracts, management has no single dashboard showing which payments have arrived and which are at risk of going overdue, so the AR picture stays fragmented across people and files.

In markets such as Thailand, where cheques remain common for business-to-business payments, missing technical interfaces make the lag worse:

• No bank API for cheque status. Most banks do not offer an automated way to query whether a cheque has cleared.
• A settlement delay. A cheque typically takes one to two business days to clear, so the "paid" status is always behind reality.
• Incomplete statement data. Finance often has to read bank statements manually, and some banks' transaction lines don't show the payer's name — which makes it hard to match an incoming amount to a specific customer.

For context, Thailand's retail payments have digitised quickly: the country's PromptPay system passed 90 million registrations and handles more than 74 million transactions a day as of mid-2025 (Bank of Thailand data, reported by The Nation). B2B settlement, however, still leans on cheques and bank-transfer slips; the national B2B rail, PromptBiz, is still in early adoption. So the reconciliation gap is a present-day operational problem, not a legacy edge case.

The result is a familiar standoff: sales believes the customer has paid, finance sees no matching deposit, and neither side can move collection forward until the two records agree.

What causes the "business–finance data disconnect"?

The disconnect comes from a structural split between the business flow and the cash flow. Sales owns the contract; finance owns invoicing and collection; the two run as separate data islands. When a deal changes or a customer delays, finance doesn't learn of it in time, which raises both the cost and the error rate of reconciliation.

A second, less obvious cause is the mismatch between real repayment rules and default software behaviour:

• Payment-allocation order. Most off-the-shelf systems default to applying a payment to the earliest overdue receivable. In financial leasing and hire-purchase, combined with cheque settlement, the common rule is to apply it to the current period first, because the cheque-clearing date is tied to the contract's current milestone. A wrong default quietly corrupts the ledger.

• Partial payments and penalty interest. When a customer pays only part of an amount, how overdue interest is calculated depends directly on the allocation order. This is where business and finance figures most often diverge.

  What are the risks of managing collections with manual ledgers and Excel?

  Managing complex payment cycles by hand scales poorly and carries measurable financial risk. Three problems recur:

  1. Lost follow-up and late collection. When a sales rep leaves or changes roles, their follow-up milestones leave with them. Without a systematic collection process, overdue accounts are often noticed only after they are seriously late — by which point a recoverable balance can turn into bad debt.
  2. No dynamic recalculation. Penalty interest on overdue instalments accrues daily. Backfilling a late payment also moves the "overdue reset" point, which then requires recalculating the earlier interest. A static spreadsheet cannot track this history reliably.
  3. Slower cash flow and internal friction. Delayed recovery constrains management's decisions, and unclear ownership pushes sales and finance into blaming each other rather than resolving the account.

  For a sense of direction, cheque use in North American B2B payments fell from 81% in 2004 to 26% in 2025 (Association for Financial Professionals survey, via Nacha). The tools are moving toward automation everywhere; firms still on manual ledgers carry the risk that trend implies.

  How can collection management be rebuilt systematically?

  The practical answer is to connect the business flow and cash flow at the data layer, then automate the parts that spreadsheets handle badly. YANGSHU approaches this through three steps for trading, manufacturing, and leasing clients.

  1. Link contracts and receivables at the data layer. Instead of scattered records, each receivable is bound to specific contract terms. Because the underlying data model keeps this link, penalty-interest calculations are traceable — the system can reconstruct the exact interest figure for any past date, which gives finance a defensible audit position during reconciliation.

  2. Generate the repayment schedule and allocation logic automatically. From the contract's interest rate, payment terms, and down payment, the system builds a repayment calendar that lays out each instalment's date and amount. Custom business logic then calculates overdue interest, handles partial payments, and applies the payment-allocation order the industry actually uses (for example, current-period-first) — reducing the manual steps where errors usually enter.

  3. Add tiered overdue alerts and a collection workflow. The system prompts the team to follow up before a due date, raises an alert and updates the dashboard once an account is overdue, and pushes the relevant contract and customer details into a collection view — shortening the time from "overdue" to "action taken."