Operations

Process mining alternative: costing your operations without event logs

Process mining is a strong tool with a hard prerequisite: clean event logs out of your source systems. If your operation runs across spreadsheets, shared inboxes and several entities on different systems, that prerequisite is the whole problem.

Last updated: July 2026

The short answer

A process mining alternative is any way of seeing and costing how work actually happens that does not depend on event logs. Instead of reconstructing processes from system data, you capture them from the people who run them, attach hours and a loaded rate to each step, and get a cost per process and per entity. It is less precise than observed event data, and it covers ground event data never reaches.

What process mining actually requires

Process mining reconstructs a process by reading the trail your systems leave behind. That definition carries three requirements, and they are where most mid-market projects stall.

Event logs

Every step needs a case ID, an activity name and a timestamp, emitted consistently by the system that handled it. Processes that run through email, spreadsheets or a phone call produce no such record.

Extraction and integration

Those logs have to be pulled out of the source systems, cleaned and joined. That is a data engineering project, and it repeats for every system and every entity in scope.

Time

Between scoping, extraction, modelling and interpretation, a first meaningful result is typically measured in months rather than days.

When process mining is the right tool

It is worth being straight about this, because the answer is often yes. Process mining is excellent when the conditions suit it:

High-volume, repetitive processes — order to cash, procure to pay, claims handling.
Work that runs end to end inside one or two systems that emit reliable event data.
An organisation with the data engineering capacity to extract and maintain the pipelines.
A question about variance and conformance: how many ways is this process actually run, and which deviate?

If that describes your operation, use process mining. Nothing below is an argument against it on its own ground.

When you need a process mining alternative

The conditions above describe large, system-mature, single-entity operations. A great deal of real business does not look like that:

The work happens outside systems — in spreadsheets, shared inboxes, phone calls and judgement. There is no event log because there was never an event.
You run several entities on different systems, so there is no single trail to follow and integration cost multiplies per entity.
You need an answer this quarter, to a board or an investor, not after a six-month data project.
Your question is what does this cost us, not how many variants of this process exist.

The last point matters more than it looks. Process mining can only see what your systems recorded. In most mid-market operations the expensive part — the rework, the chasing, the manual reconciliation between two entities' systems — is precisely the part that leaves no trace.

How Espresso approaches the same question

Espresso captures the process from the people who run it, then costs it directly. No connectors, no event logs, no IT project. Four steps:

Step 01

Map

Lay out each process across each entity. Templates seed a full board, so you start from what your team already knows rather than a blank page.

Step 02

Cost

Hours per week against a loaded rate, plus leakage. You get an annual cost per process, per entity, and rolled up across the group.

Step 03

Diagnose

Crema reads the pain points, systems and target architecture on a process, then proposes a re-engineering path with best-practice TO-BE flows.

Step 04

Deliver

Push the diagnosis into a delivery workspace — tasks, owners, timelines — tracked against the process it came from.

What you give up

This is a trade, not a free win, and you should go in knowing which side of it you are taking.

You give up observed precision

Espresso's inputs are estimates from the people doing the work. Event logs are a record of what happened. An estimate of eight hours a week is not the same class of evidence as eight hours measured, and it should not be presented as one.

You gain coverage and speed

In exchange you see the work that never touched a system, across every entity, in a working session rather than a project. For most operations the map is directionally right and immediately actionable, which is usually what the decision needs.

When precision is the point, mine

If you are proving conformance to a regulator or optimising a process worth millions at the margin, the accuracy of event data is worth its cost. Use the right tool.

Common questions

What is a process mining alternative?

Any method of understanding and costing how your operations actually run that does not depend on extracting event logs from source systems. The practical alternatives are consultant-led process mapping, internal workshops, or a structured tool that captures the process from the people who perform it and costs it directly.

Can you do process mining without event logs?

Not in the strict sense — process mining is defined by reconstructing process flows from system event data, so without logs there is nothing to mine. What you can do is capture the same picture a different way: ask the people who run each process how it works, how long it takes, and where it breaks, then cost that directly.

Why is process mining hard for mid-market companies?

It assumes your processes are captured end to end inside systems that emit usable event logs, and that you have the data engineering capacity to extract them. Mid-market operations often run across spreadsheets, email and phone calls, and across several entities on different systems, so much of the real work leaves no event trail to mine.

What does Espresso cost an operation on?

Hours per week against a loaded hourly rate for each process component, plus estimated leakage from rework and errors. That produces an annual cost per process, per entity, and rolled up across the portfolio.

Cost of operations — the method for putting a real number on a process without event logs.
What is an operations OS? — the category, and how it sits against process mining.
For holding companies — the same question, asked across several entities at once.

See what your operation costs

Map your processes, attach real hours and rates, and get a cost per process and per entity — without an integration project in front of it.

See how Espresso works

Handling sensitive operational data? Read how we secure it.