What Is Process Mining?

What Is Process Mining?

A plain-English guide for CFOs, Finance Directors, and Programme Leaders involved in ERP transformation.

Overview of process mining

Process mining is a research area and technology that helps businesses understand their real processes, how they're operated, and identify opportunities for improvement, automation, and digitalisation.

Process mining works with your existing systems and data to provide end-to-end visibility and enable faster, more informed business decisions. By extracting event log data from your systems of record — the timestamped history of every activity your processes have performed — it reconstructs and visualises how processes are actually executing across your organisation. This allows you to compare actual process behaviour against the designed process, identify variants and exceptions, determine the root cause of inefficiencies, and monitor performance against KPIs.

Overall, process mining is a valuable tool for businesses seeking to improve their operational efficiency and make informed decisions.

What is process intelligence?

Process intelligence is the layer of insight and decision-making capability that sits above process mining. Where process mining discovers and visualises what is happening in your processes, process intelligence asks what it means and what you should do about it. It combines process mining data with advanced analytics, AI, and business context to move beyond observation into action — enabling organisations to model the impact of proposed changes before implementing them, predict where process failures are likely to occur, and continuously monitor improvement over time. The result is not just a map of how your processes behave today, but a dynamic, evidence-based foundation for how they should perform tomorrow.

How does it compare to process mapping?

Traditional process mapping relies on interviews, workshops, and observation — it captures how people believe a process works, documented at a point in time by the people involved in running it. Process mining, by contrast, reads directly from the data your systems generate, reconstructing how processes actually execute across every transaction, every day. Where a process map reflects a consensus view that may be incomplete, idealised, or quickly outdated, process mining reflects reality — including the variants, exceptions, and workarounds that never make it onto a whiteboard. The two are not mutually exclusive; process maps remain useful for communication and design. But process mining provides the evidence base that process mapping alone never can.

How does it apply to finance?

Finance functions are among the most natural candidates for process mining because they run high-volume, rule-based processes that generate rich transactional data as a matter of course. Every invoice, payment, journal entry, and reconciliation leaves a timestamped trail in your systems of record — and that trail is exactly what process mining needs.

The three core finance process cycles are where the value is most immediately visible.

Purchase-to-Pay

Process mining reveals where invoices are stalling, which suppliers consistently trigger query loops, and where approval flows are adding delay without adding control.

Order-to-Cash

Process mining surfaces the gap between your designed order-to-revenue cycle and what is actually happening across your customer base.

Record-to-Report

Process mining identifies what is driving month-end overrun, where manual interventions are concentrated, and which journals are generating rework.

For Finance Directors and CFOs, the significance is not just operational — it is strategic. Process mining provides the evidence base to answer the question that most transformations leave unanswered: the system is live, so where are the promised efficiency gains?

How does it connect to the data source?

Process mining connects to your data through an event log — a structured extract from your systems of record containing three essential elements: a case identifier (such as an invoice number, order ID, or journal reference), an activity name (the step that occurred, such as "invoice received" or "payment approved"), and a timestamp recording exactly when that step took place. Optional attributes — such as the resource who performed the activity, department, or transaction value — can be added to enrich the analysis. Most modern ERP and finance systems, including Microsoft Dynamics 365, SAP, and Oracle, hold this data already; it is a question of extracting and structuring it correctly rather than creating anything new.

Starting simple — CSV and text file imports

The simplest way to get started is a CSV or text file export from your finance system. This is a low-friction first step that requires no integration work and can produce a first process map within hours. It is deliberately designed as a quick win — giving Finance leaders an immediate, evidence-based view of their processes before committing to anything deeper.

Moving to direct connection

For ongoing analysis, direct connectors to source systems replace the manual export, keeping the process map continuously updated as new transactions occur. This turns a one-off diagnostic into a live process intelligence capability — a real-time view of how your finance function is actually performing, refreshed automatically as your data changes.

Is our data safe?

Data security is one of the first questions we hear — and it is a reasonable one. It is worth being clear about what process mining actually uses, because it is often less sensitive than people assume.

What data does process mining use?

Process mining works with event log data — the timestamped record of process activity generated by your systems. This means activity names, timestamps, case identifiers, and process steps. It does not require access to personal data, payroll records, or commercially sensitive financial figures. In most finance implementations, the event log can be scoped to exclude anything beyond what is needed for process analysis.

Where does the data go?

Process mining can be deployed within your own environment, meaning your data never leaves your infrastructure. Where cloud-based analysis is used, data is processed within enterprise-grade platforms that meet recognised security and compliance standards. The specific architecture — on-premise, cloud, or hybrid — can be agreed before any engagement begins, and we are happy to work within your organisation's existing data governance framework.

Who has access?

Access to process mining outputs is controlled and role-based. The process maps and analytics generated are shared only with the stakeholders you designate — they are not visible to third parties, and raw data is not retained beyond the scope of the engagement.

Do we have to buy software?

No — and this is one of the most common misconceptions about process mining. Many organisations assume that getting started requires a significant software investment, a lengthy procurement process, and a dedicated technical team to manage the tooling. In practice, none of that is necessary to see value quickly.

Starting with a diagnostic

Tierpoint's Finance Process Diagnostic requires no software purchase on your part. We bring the tooling, run the analysis, and deliver the output. Your investment is time — specifically, the time needed to produce a structured extract of your transaction data — and the diagnostic engagement itself. There is no licence commitment, no infrastructure requirement, and no obligation to proceed further.

If you want an ongoing capability

For organisations that want to move beyond a one-off diagnostic to a continuous process intelligence capability, software will eventually form part of the picture. When that point arrives, we will recommend the right tooling based on your existing technology estate, your team's capabilities, and the scale of what you want to achieve. If you are already a Microsoft customer, there are options that sit natively within your existing licences that may reduce or eliminate additional cost entirely.

The principle

Our view is simple: you should see the value before you commit to the investment. The diagnostic is designed to do exactly that.

How long does it take?

The honest answer is: less time than most people expect — and it scales with what you want to achieve.

Getting to a first process map

With a CSV or text file export from your finance system, a first process map can be generated within hours of receiving the data. The data extraction itself — pulling the event log from your ERP — typically takes your finance or IT team a matter of hours rather than days. From the point of receiving a clean data extract, Tierpoint can have an initial process map and a first layer of analysis in front of your team within days.

The full diagnostic

Tierpoint's Finance Process Diagnostic — covering Purchase-to-Pay, Order-to-Cash, and Record-to-Report — is a fixed-scope engagement typically completed within two to three weeks from data receipt to final output. That includes the process maps, variant analysis, root cause findings, and a prioritised view of improvement opportunities. There is no open-ended discovery phase and no scope creep — the deliverable and the timeline are agreed upfront.

Building an ongoing capability

For organisations that want to move to a continuous process intelligence capability with live data connections, the timeline will depend on the complexity of your systems landscape and the scope of what you want to monitor. That said, pre-built connectors for common ERP platforms mean that direct data connection is typically measured in days rather than weeks.

The principle

Process mining should show you something valuable quickly. If it is not doing that, something is wrong with the approach — not the technology.

How much does it cost?

Process mining engagements vary in cost depending on scope and whether you are running a one-off diagnostic or building an ongoing capability. We are deliberately transparent about how we structure this.

The diagnostic

Tierpoint's Finance Process Diagnostic is fixed-price at £8,000 per process. You will know the cost before you commit, and it will not change. Each process diagnostic — whether Purchase-to-Pay, Order-to-Cash, or Record-to-Report — delivers a complete process intelligence report with quantified improvement opportunities, variant analysis, and root cause findings. Run all three for a complete view of your finance function, or start with the process where you suspect the most value is trapped.

Ongoing process intelligence

For organisations that want to move beyond the diagnostic to a continuous process intelligence capability, cost will depend on the tooling selected, the number of processes in scope, and the level of ongoing support required. We will always provide a clear commercial proposal before any work begins, and we will only recommend tooling investment where the expected return justifies it.

The return

The more relevant question for most Finance Directors is not what process mining costs — it is what the absence of it is costing. Inefficient P2P processes, extended month-end cycles, and unresolved workarounds in an ERP that has already been paid for all carry a quantifiable cost. The diagnostic is designed to make that cost visible — and to make the case for improvement on your own data, not ours.

How do I get started?

Getting started is simpler than most people expect. There is no procurement process, no software installation, and no lengthy scoping exercise before you see something valuable. If you can export a spreadsheet, you can get started today.

Step 1 — Pick a process

Decide where you want to start. For most organisations, Purchase-to-Pay is the natural first choice — it is high volume, well-understood, and almost always contains more inefficiency than the finance team realises. If month-end overrun is the burning issue, Record-to-Report may be the better starting point. We are happy to discuss which process is likely to yield the most immediate insight based on your current situation.

Step 2 — Export six months of transaction data as a CSV

Ask your finance or IT team to export the last six months of transaction activity from your finance system as a CSV file. That is it. The file needs three columns: a case identifier (such as an invoice number or journal reference), an activity name (the step that occurred), and a timestamp (when it occurred). Most ERP systems can produce this in minutes — no specialist knowledge required.

Step 3 — We take it from there

Share the CSV with us and we handle everything from that point — data ingestion, mapping, transformation, and analysis. If at any stage you want to move to a live, continuously refreshed connection directly from your source system, we can establish that too. But a CSV is all you need to get started.

Step 4 — Your process map, within days

Within days of receiving a clean data extract, we will have a first process map in front of you. From there, the full diagnostic — covering root cause analysis, variant comparison, and quantified improvement opportunities — is typically delivered within two to three weeks.

Tierpoint Partners is a specialist Microsoft D365 Finance Transformation advisory firm. We work with CFOs and Finance Directors at mid-market organisations to de-risk transformation programmes and recover value from underperforming finance functions.