MDM Studio unifies customer, product, party and any other domain data scattered across your systems into governed golden records — the whole method in one product, running on the database platform you already own.
Every large business runs on records that disagree. The core system, the CRM, the billing platform and the branch spreadsheet each hold a version of the same customer, product or supplier — and none of them is wrong enough to fix, or right enough to trust.
Duplicate customers inflate campaign spend and distort every count. Reconciliation happens in spreadsheets. Analytics is argued with rather than acted on.
A correction made in one system does not reach the others. When a regulator or auditor asks who changed what and when, the answer has to be assembled by hand.
Every personalisation or single-view initiative starts by rebuilding the same customer view from scratch — and inherits the same disagreement.
MDM Studio fixes the record once, in a governed hub, and keeps it fixed — then serves the corrected value to everything downstream and, where you allow it, writes it back to the systems that hold it.
MDM Studio walks your master data through the full lifecycle — from a blank canvas to golden records the whole business can rely on. There are no modules to license one by one.
Model the estate visually: domains, entities, a governed data dictionary, relationships and ERDs.
Connect SQL Server, Oracle, CSV, Excel, FTP and REST sources. Map fields stage by stage and let scheduled pipelines do the heavy lifting.
Change requests with review and approval, policies, role-based access, sensitive-data clearance and full audit trails — built in, not bolted on.
Match, merge and survivorship produce golden records with complete cross-reference lineage, quality scoring and controlled write-back.
MDM Studio hub works identically on all four supported database engines.
| Office | The question you are asked | What MDM Studio gives you |
|---|---|---|
| Chief Executive & the Board |
“Do we actually know how many customers we have?” | One governed count with a stated basis and confidence, visible on a dashboard rather than assembled for a meeting. King V Principle 9 makes data integrity a board responsibility from 1 January 2026 — this is the evidence for it. |
| Chief Risk & Compliance |
“Prove this correction reached every system.” | A tamper-evident audit trail of who changed what and when, approval records on every published change, and governed write-back that pushes the corrected value to source under eligibility rules and post-apply verification. |
| Chief Financial Officer |
“What is duplication costing us?” | Duplicate detection with scored, explained matches; supplier and product mastering that collapses regional vocabularies through crosswalks; and quality scorecards that turn “the data is bad” into a number with a trend. |
| Chief Information & Data Officer |
“Can we do this without a two-year programme?” | One product covering design, mapping, quality, mastering and governance, on the database platform you already run, deployed inside your own perimeter — starting on a single VM. |
Domains, entities and attributes are unlimited at every tier. Modelling a second or a tenth domain costs nothing, so the programme grows by appetite rather than by budget round.
One golden party across core banking, CRM, cards and policy administration, with classification and masking on personal fields, and an approval record behind every published change. Supports FICA client-due-diligence completeness and the accuracy duty POPIA places on personal information.
Product and supplier mastering across regions and banners, with reference-data crosswalks that reconcile competing code sets rather than forcing one on everybody. Golden values served to downstream systems through a read-only API with revocable tokens.
Material, equipment and vendor masters with hierarchy tiers, per-tier quality scoring and governed reference vocabularies — so spend analysis and maintenance planning read the same catalogue.
One canonical model, natively on SQL Server, PostgreSQL, MySQL and Oracle. The engine is a per-session choice, not a migration project.
Governed code sets with immutable, checksummed releases and crosswalks between source vocabularies.
Push mastered values back to source under eligibility rules, batching, approval gates and post-apply verification.
Many isolated business units or subsidiaries on one deployment, enforced at the data layer — no shared record pool.
Multi-level hierarchies with per-tier dimensions, labels and their own data-quality scoring.
A report studio in the hub — seven visual types, calculated fields, pivots with drill, dashboards whose filters unify across sources, saved views and scheduled subscriptions — plus end-to-end lineage views. Reports read through the same clearance and row-level rules as the record screen, so analytics is not a way around them.
Point it at SQL Server, PostgreSQL, MySQL or Oracle. Migrations bootstrap the schema on first run. Automatic HTTPS via Caddy and Let's Encrypt.
Many connection types across relational, cloud, file and integration sources, in one registry with credentials AES-256 encrypted at rest. Schema discovery and live previews from your real sources.
A guided pipeline takes a domain from first extract to golden record. Advisors propose match, survivorship, quality and crosswalk configurations for a steward to accept or reject.
The Adaptive Canvas Academy is free at every tier. Four certification tracks with graded exams and verifiable credentials valid 24 months: Certified Associate (Foundations), Certified Data Steward, Certified Modeler & Integrator and Certified Administrator. Pass mark 70%. A guided lab builds a complete Customer 360 in a real environment in about 150 minutes. Certificates carry an integrity-checked ID anyone can verify — which also means you can require it of an implementation partner.
Every deployment exports a one-click support bundle with secrets automatically redacted, and exposes live health diagnostics and a Prometheus metrics endpoint — no server login required. A conversation starts with data, not guesswork.
| Severity | Standard | Premium 24×7 | Resolution objective |
|---|---|---|---|
| S1 — Critical Production down; no workaround |
4 business hours | 1 hour, 24×7 | Continuous effort until restored |
| S2 — High Major function impaired |
1 business day | 4 hours | Fix or workaround within days |
| S3 — Normal Minor issue; workaround exists |
2 business days | 1 business day | Scheduled into a maintenance release |
| S4 — Low Cosmetic, docs or feature request |
Next business day ack. | Next business day ack. | Considered for the roadmap |
First-response targets. These are objectives, not contractual guarantees; exact commitments are defined in your support agreement.
MDM Studio installs on-premises, in your private cloud, or as an isolated tenant on a managed hub. Wherever it runs, the hub sits next to the databases it masters and each tenant is isolated at the data layer, with no third-party or SaaS egress — nothing is shipped to a vendor cloud to be processed. POPIA does not require data to stay in South Africa; section 72 sets conditions for transferring it. Knowing exactly where your master data sits, and being able to prove it, is what makes that a question you can answer.
Four sensitivity levels cascade from domain to record; clearance governs access; values are masked at serve time; the audit trail proves it.
Local password, Active Directory/LDAP or OIDC single sign-on, chosen per user, with time-based MFA. SCIM and SAML at Enterprise.
Connection and directory secrets AES-256 encrypted at rest, decrypted only in memory, never returned to the browser.
Capability picks the edition; mastered-record volume prices it. You are never charged per domain, per module or per attribute, and never per record — a band is pre-paid capacity, not a meter running against you. Licences are signed keys, and if a key lapses MDM Studio falls back to Community rather than locking you out.
Choose the edition for what it does — hub engines, sign-on, write-back, real-time serving, HA/DR — and the band for how much you master. The two are independent, which is why the same volume appears under more than one edition.
| Mastered records | Community | Professional | Enterprise | Enterprise+ |
|---|---|---|---|---|
| Up to 25,000 | Included · R0 | — | — | — |
| Up to 100,000 | — | R450,000 | — | — |
| Up to 250,000 | — | R625,000 | — | — |
| Up to 500,000 | — | R795,000 | — | — |
| Up to 1,000,000 | — | R1,050,000 | R1,550,000 | — |
| Up to 2,500,000 | — | — | R2,100,000 | — |
| Up to 5,000,000 | — | — | R2,850,000 | — |
| Up to 10,000,000 | — | — | — | R3,600,000 |
| Up to 25,000,000 | — | — | — | R4,800,000 |
| Above 25,000,000 | — | — | — | From R6,500,000 |
Indicative annual licence in ZAR, self-managed, confirmed by sizing. Effective cost falls from about R4.50 per record at the entry band to R0.57 at five million and R0.19 at twenty-five million — the ladder rewards growth rather than taxing it. Where a volume appears under two editions, the difference between them is what the higher edition does, not how much you master.
| Cost line | How it behaves |
|---|---|
| Licence | Annual, banded by mastered record volume. Unlimited domains on every paid edition, entities and attributes — modelling another domain costs nothing. |
| Hub database | R0 if you use PostgreSQL or MySQL. Prefer SQL Server or Oracle? You are using licences you almost certainly already own. |
| Infrastructure | One VM to start; the hub sits next to the databases it masters. No egress charges for shipping master data to a vendor cloud, because it is not shipped. |
| Implementation | Scoped by domain and source count, delivered by us or by a certified partner. |
| Training | R0. The Academy and its four certifications are free at every tier. |
| Support | Standard business-hours support and guided onboarding included with Professional and Enterprise. Premium 24×7 optional. |
Indicative annual licence in ZAR, confirmed by sizing. 3-year prepaid attracts a term discount. Vendor-hosted deployment is priced separately from self-managed. This is not an offer.
Data sovereignty and self-hosting; four hub engines; the whole method in one product, reporting included, so answering a question about mastered data does not mean exporting it somewhere less governed; licensing you can read; speed to a first governed domain.
Established suites bring deeper capability breadth, packaged industry data models, large partner ecosystems and analyst coverage.
Our partner bench, reference base and certified community are growing. We would rather say so than oversell it.
Performance: the matching engine and pipeline are benchmarked in an in-product Performance Lab on generated data — a representative envelope you can reproduce, not a service level. The largest volume tested is 5 million records. Figures are available on request and we will not extrapolate beyond what we have measured.
We will connect a sample of your own sources and walk your data from first extract to golden record, live — on a hub standing on one VM, inside your perimeter. You will see the match evidence, the survivorship decisions and the audit trail on your own records, and leave with a defensible estimate for the full estate.
info@adaptivecanvas.co.za · adaptivecanvas.co.za · Partner enquiries: partners@adaptivecanvas.co.za
MDM Studio v1.2.0 · © 2026 Adaptive Canvas · adaptivecanvas.co.za · South Africa