Skip to main content

CryspIQ® vs Inmon

These two agree on the goal and disagree on the cost of reaching it. Inmon and CryspIQ® both aim at one enterprise model with consistent definitions across every function — and both, unlike most alternatives, actually deliver it. The difference is that under Inmon you design and build that model before it returns anything, and under CryspIQ® the structure is already built, patented and ready on day one, so the work is mapping your sources into it.

What Inmon is​

The Inmon method is the top-down approach: build a normalised, subject-oriented enterprise data warehouse first — the Corporate Information Factory — as the single integrated store of the business, then derive departmental data marts from it.

The reasoning is sound and has aged well. If every mart descends from one integrated model, the marts cannot disagree, because there is only one definition upstream of all of them.

Where the two actually differ​

When the value arrives. An Inmon programme front-loads the modelling: the enterprise model has to be designed, agreed across departments and populated before much of anything is queryable. That is where the reputation for long, expensive warehouse projects comes from — not from the method being wrong, but from the whole cost landing before any of the benefit. CryspIQ® inverts it: the structure ships with the product, so the first mapped source is queryable while the rest of the estate is still being connected.

Who settles the definitions. Inmon's integrated model requires the organisation to agree what a customer is before the model can be built — and that argument, not the technology, is where most enterprise data programmes actually stall. CryspIQ® moves the question: the structure already says a party is a party and a monetary fact is a monetary fact, so the discussion becomes mapping each source onto a definition that already exists, rather than negotiating a new one from nothing.

What a change costs. Adding a subject area to an Inmon warehouse touches the enterprise model and everything derived from it, which is why change impact is high and gets higher as the estate grows. In CryspIQ® the structure does not move when the business does — a new subject area is mapping and reference data.

Lineage. Inmon lineage is maintained by whoever builds the pipelines. CryspIQ® traces a figure back to its source record through the link key, without anyone documenting it.

Where Inmon is stronger​

Design freedom, and the fact that it assumes nothing about your business. CryspIQ®'s structure is fixed at the level where organisations are genuinely alike — every business records money, quantities, references, events, positions, rates and documents — and open at the level where they differ. If your business needs a structure outside that, Inmon lets you build it and CryspIQ® does not.

A mature, well-governed Inmon warehouse that already exists is also a poor thing to throw away. The argument for CryspIQ® in that situation is about what it costs to keep changing, not about whether it works.

Using both​

CryspIQ® reads the raw or staging layer you already populate, so it does not require the existing warehouse to be decommissioned or migrated — both run at once, fed from the same place, and layers are retired only when nothing reads them. See co-existence.