Since March 2020, finance and accounting teams across Turkey have been working from home. Month-end close was already a pressure point before the pandemic; the informal coordination of a shared office — a quick word across the desk, a glance at a colleague’s screen — provided a safety net that nobody had formally documented. When that disappeared overnight, the cracks widened. Phone traffic increased, email chains grew longer, and the question of who was waiting on whom became genuinely unclear. Completing the monthly financial close on schedule turned into a real operational problem for many companies. The traditional response to this problem remains unchanged: meetings, surveys, a manager’s observation. Process mining approaches the same problem from a different direction entirely.
Process mining is a discipline that analyses the event logs produced by enterprise systems — ERP, accounting software, approval workflows — to make visible how a process actually runs, as opposed to how it is assumed to run. The gap between the process map in a manager’s head and the sequence of events recorded in the system is often surprisingly large. In the context of financial close, this kind of analysis is particularly useful because the close process consists of dozens of sub-steps: posting accruals, completing bank reconciliations, eliminating intercompany transactions, routing approvals, consolidating entities, and producing reports. Which of these steps is causing delays, at what time of day the bottleneck concentrates, which user action or system dependency is the root cause — all of this is recorded in the event log. The challenge is converting that record into a readable, actionable form.
Consider a mid-sized manufacturing or retail company in Turkey running SAP, Logo, or Netsis as its ERP backbone, with an accounting team of between five and fifteen people, and a month-end close that normally completes somewhere between the fourth and eighth business day of the following month. During the pandemic period, that timeline stretched by two to four additional days for many such companies. To locate the source of the delay, a process mining tool connects to the ERP event logs and extracts every document movement, every approval step, every status change — each with a precise timestamp. The resulting analysis typically reveals that sixty to seventy percent of total delay is concentrated in two or three critical steps. The remaining steps complete more or less on time. This finding frequently contradicts management intuition, because intuition tends to point at the most visible problem rather than the most consequential one.
Identifying delay points with data has a direct effect on reporting reliability. As the close cycle extends, two things happen: reports reach the board or investors later, slowing decision-making; and last-minute entries and corrections multiply, increasing the probability of errors. In Turkey, publicly listed companies are required to submit financial statements to the KAP (Public Disclosure Platform) within fixed regulatory deadlines — a delayed close translates directly into legal compliance risk. For SMEs, bank credit assessments and supplier payment schedules frequently depend on periodic financial statements. A delayed close triggers a domino effect across these processes. When process mining identifies and removes the delay points, not only does the close timeline shorten — the number of accrual and correction entries in the final reports also falls. That is a measurable indicator of reporting quality, not just a scheduling improvement.
The most common practical obstacle in implementation is data quality. For process mining to produce reliable results, the ERP event logs must be consistent and complete. Many Turkish SMEs have been running their ERP systems for years, but system configurations have drifted over time, and certain steps have migrated outside the system entirely — handled through Excel files, email threads, or phone calls. In that situation, process mining shows only the portion of the process that the system has actually recorded, not the full picture. A data quality assessment and scope review before the analysis begins is therefore not optional. Licensing costs for process mining tools can also be a significant barrier for companies with limited IT budgets; cloud-based subscription models have reduced this barrier to some extent, but data security and compliance with Turkey’s personal data protection law (KVKK) remain separate considerations that require careful evaluation before any data is transferred to an external platform.
For an SME finance director weighing whether to use process mining for the financial close, the decision criteria should be concrete. Are your ERP event logs accessible and sufficiently granular? How many steps in your close process are recorded in the system, and how many are handled outside it? Does the cost of delay — late reporting, management time lost, error correction — justify the investment in tooling? If you can answer those three questions clearly, process mining gives you an evidence base for improvement rather than a set of assumptions. If you cannot, strengthening the data infrastructure first is a more productive starting point. For finance teams now working in a distributed setup under pandemic conditions, getting clear answers to these questions matters more than it did a year ago — because the informal visibility that used to compensate for process gaps no longer exists.
This article was originally written in Turkish by Gökhan MERCANOĞLU on May 25, 2020 and has been automatically translated into English and other languages using machine translation.