Master Data · Data Quality Available

Find where your master data is wrong, and fix what matters first

Rules by dimension, every failing record as a finding, and a weighted score per entity. A profiler shows how your data is filled, and AI suggests rules where coverage is missing.

Become a founding customer Take the tour

Data QualityAvailable

86%Items

91%Customers

78%Suppliers

Items, by dimension

Completeness92%
Validity88%
Consistency81%
Uniqueness99%
Plausibility84%

Top findings

Stocked items without a receiving location37
Item/warehouses with a blocked supplier9

Example scores and findings.

Rules

Five dimensions of quality

Every rule belongs to a dimension, so you see not only that data is wrong, but in what way.

CompletenessThe data a process needs is there.e.g. Stocked items have a receiving location

ValidityValues are allowed and usable.e.g. The supplier on an item/warehouse record is active, not blocked

ConsistencyValues agree with each other.e.g. The reorder point is not below the safety stock

UniquenessNo duplicates where there should be one.e.g. No two customers with the same VAT number

PlausibilityValues make sense.e.g. A lead time between 0 and 180 days

A tour of the screens

From the score to the record, in five screens

Dashboard: the score per entity, the weighted coverage and the open findings
Coverage: which fields have rules, per dimension, and where the gaps are
Rules: every rule with its dimension, active or not, and its findings
Findings: the records that break a rule, with the field and the value
Profiler: how each field is filled, before you write rules

M3 Data › Data Quality · item/warehouse

84%Overall score

71%Weighted coverage

245Open findings

Score per entity

Items86%49 findings
Item/warehouse82%112 findings
Customers91%23 findings
Suppliers78%61 findings

Example screens, entities and figures.

Rule generation

Generate rules, instead of writing them all by hand

Start from your data: profile it, let AI suggest what is missing, and decide which rules to switch on.

1ProfileSee how each field of a table is filled: how often it is empty and which values occur.
2SuggestWhere a field has no rule yet, AI proposes one, based on the field and how it is filled.
3ReviewYou decide which suggestions become rules, and which you dismiss.
4ActivateSwitch the rule on. From then on it counts in the findings, the score and the coverage.

Profiler · item/warehouse

Field	Filled	Distinct	Most common
WHSL · Location	96%	1,240	A-01 (8%)
ALOC · Allocatable	100%	2	1 (94%)
RESP · Responsible	61%	14	(empty) (39%)
LEAT · Lead time	88%	57	14 (31%)

✦ AI suggestion · completeness

Responsible (RESP) is filled for every active item/warehouse

Based on the profile: 39% of records have no responsible person today.

Add ruleDismiss

Example profile and suggestion.

Coverage

See where rules are missing

A matrix of fields against dimensions. At a glance you see what is covered, which rules are switched off, where a rule is missing, and where AI suggests one.

Coverage counts in the score: an entity with few rules can’t look good just because nothing is checked.

Coverage · item/warehouse

Field	Complete	Valid	Consistent	Plausible	Unique
WHSL Location	✓	•	–	–	–
SUNO Supplier	•	✓	✦	–	–
REOP Reorder point	•	•	✓	✦	–
LEAT Lead time	•	•	–	‖	–

✓Covered•Gap‖Inactive rule✦AI suggestion–Not applicable

Example fields.

Making the rules count

Find it, fix it, and stop it coming back

1Active rules find the recordsEvery record that breaks an active rule becomes a finding, with the field and the value that is wrong.

2Critical fields weigh moreFields under governance on L2 or L3 count as critical, so the score reflects the data that steers M3.

3Fix, then check againCorrect the records, for example in Excel with Data Export and Data Import, and see the score move.

4Stop it at the sourcePut the field under Data Governance, so the next change to it is seen, and signed off where it matters.

Data Import →Data Governance →

What it brings you

Master data you can rely on

1Fix what matters firstFields under governance count as critical, so the score reflects what steers M3.
2From score to recordEvery number leads to the records behind it.
3Rules where you need themThe coverage view shows which fields have no rule yet, and AI suggests one.
4Ready for go-liveKnow whether the data is ready before a migration or a new process.

Next in your journey

Data GovernanceAvailableKeep the data right once it is fixed.“Who changed this, and was it allowed?” →Data ImportAvailableLoad the corrections safely, with a backup.“Can I load this safely?” →Field ProvenanceAvailableFind where a wrong value comes from.“Where can this value come from?” →

Coming next Roadmap

Rules that know which process they protect

With the knowledge graph, each rule will show which process step it protects, what goes wrong when it fails, and how often that process runs in your data.

Make your master data a reason things work

We are onboarding a small group of founding customers. You get the newest features first, a direct line to the team, and founding pricing.

Become a founding customer

Preview features firstPredictor, Debugger, Release Toggle Impact and Value Provenance
Shape the roadmapYour M3 questions decide what we build next
Founding pricing, locked inYour rate stays as the product grows