Test the full population. Not a hundred cases, but a hundred thousand.
DIVE reconstructs from an export of your source system how a process actually ran, sets that against the design in Flowmap and turns the deviation into an audit finding. Every control test runs on the full population, with the coverage and a frozen definition per test run recorded. No sample that misses the exception.
The challenge
A sample usually does not find the exception
Operational systems already record every step: who, when, in what order. That data goes largely unused, and the audit falls back on a manual selection of twenty-five or forty files.
Twenty-five files say nothing about ten thousand
A control that was bypassed in 591 of 10,000 cases usually goes unnoticed in a sample of forty files. The exception you are looking for is exactly what a sample is statistically likely to miss.
Two tools for one question
A generic process mining platform draws how the process ran, but stops at the picture. Translating that into one control, one norm and one finding remains manual work in a second tool.
Last month is no evidence for this month
An analysis you ran last quarter is no answer to the question of what was tested this quarter. Without recorded repetition, every round starts with a reconstruction of what was actually done last time.
A changed threshold overwrites the old result
If you adjust a mandate limit or a threshold, the March result should remain attached to the March definition. Without version control on the test itself, the audit file silently shifts to the new limit.
Segregation of duties on paper, not on what actually happened
An access rights analysis sees who is allowed to. It does not see who actually did it through an emergency account, role accumulation or substitution, and that is exactly where the finding sits.
An event log with a name is an employee monitoring system
As soon as it contains a performer field, it falls under the GDPR and article 27 of the Dutch Works Councils Act (WOR) gives the works council the right of consent. Do not arrange that after the fact.
What DIVE offers
From process picture to demonstrable control test
Other process mining tools show how a process ran. DIVE sets that picture against the design and the standards framework, and turns the difference into a control test that can be explained to a third party.
Population instead of sample
Every control test runs on the full export, not on a selection of forty files. A bank account that changes shortly before a payment is almost impossible to find with a sample and a matter of seconds on the full population. Exactly where specialised audit tech such as Audit Sight and DataSnipper sets itself apart.
Every test run is a frozen audit file entry
A control test is a record with a version number. Every execution records the population size, the coverage and a frozen copy of the definition, so the March result stays attached to the March definition.
Segregation of duties on what actually happened
DIVE tests two incompatible roles against the actual execution, including emergency accounts, role accumulation and substitution. An access rights comparison does not see that.
Designed versus actual, made spatial
Link a Flowmap diagram and see in one 3D view which steps skip a control, which path was never designed and which designed path never occurs.
Traceable to GIAS Standard 14.1
Every figure carries its evidence trail: dataset, period, number of events and the rule applied. A competent third party can recalculate it and reach the same conclusion.
Privacy by design
The performer is irreversibly pseudonymised at import, what you do not map is not stored, and the raw file is removed as soon as the log is built.
One summary for the audit committee
Fixed sections, no recalculation, and the caveat always visible: what was not tested belongs to the opinion just as firmly as what was found.
The same test, the next export
A recipe stores the parameters with the source system. DIVE recognises the next export by its column names and offers a dry run before applying them, so repeating is never done blind.
How it works
From export to finding in seven steps
DIVE guides you through the whole analysis, from a raw CSV or Excel export to a validated finding with an evidence trail.
Upload the export
Drag in a CSV, TSV or Excel file, up to 10 GB. Large files go in blocks of 8 MB and can be resumed after an interrupted attempt.
Map the columns
DIVE proposes a mapping based on column names and, for the timestamp, measures every possible date format against the actual values, so you can see for yourself whether 03-04 is March or April.
Assess the data quality
Before any analysis runs, you see what is wrong: missing case IDs, unreadable timestamps, duplicates. An administration in which eight per cent of the rows has no case ID is a finding in itself.
Analyse
Process discovery, variants, throughput time and segregation of duties, all on the full population.
Test against the design
Link a Flowmap diagram and let DIVE determine where practice deviates: a control bypassed, an undesigned path, or a path that never occurs.
Validate and record
A hit is an indicator, not a finding. Confirm it against the source document, and only then does the finding come into being, with an evidence trail.
Save as a recipe
Once the analysis is set up the way you want it next time, save the parameters with the source system. Next month's export gets them back automatically.
The test run
What every execution records
A control test is a record with a name, a type, parameters and a version number. Every time you run it, a test run is created with four fixed data points.
Population size
The number of cases the test was run over: the full export, not a selection from it.
Coverage
The part of the population the test was able to form an opinion on. What falls outside it does not silently count as approved.
Frozen definition
A copy of the parameters as they applied at that moment, so the result never silently shifts under a later threshold.
Version number
If you change a threshold, the version counts up. If you only change the name, it does not: that does not shift the result.
The eight control tests
Fixed tests, no black box
Rule-based and without machine learning, so the outcome can be explained to a reviewer. Every test runs on the full population and delivers a coverage percentage alongside the exceptions.
Segregation of duties
The same person in two incompatible roles within one case, including emergency accounts and substitution.
Duplicates
Cases that share the same value on one or more attributes, the classic pattern of a double payment.
Authorisation threshold
A value above the mandate limit without the required approval step.
Master data change before payment
A master data field that changes shortly before a payment, in the same case. Almost impossible to find in a sample, a matter of seconds on the population.
Time window
Postings outside office hours, at the weekend or after the period close.
Sequence gap
Missing numbers in a continuous sequence.
Reconciliation
The count of the source data against a control total or general ledger item.
Benford
Cases with a first digit that is significantly overrepresented, judged against Nigrini's bounds.
Who it is for
For internal audit, IT audit and compliance
For teams that want to examine a process based on what actually happened, not on an interview or a sample. DIVE supports both the recurring quarterly test and the one-off deep dive after an incident.
Curious what DIVE finds in your own data?
DIVE is currently running in a closed test phase. Get in touch for a demo on your own export format.