01. Connected Common Data Environments: What They Are and What They Actually Solve.
Overview
A connected common data environment is an arrangement in which two or more Common Data Environments (CDEs) are kept in step automatically, with each platform still owned and administered by the party that licences it. Files, folder structure, version history, metadata and workflow state move between the platforms on a defined route, so a document issued in one environment appears in the other in a usable form. No tenancy is merged and no party changes platform.
Most major projects run several CDEs at once. The usual method of moving information between them is a person downloading files from one platform, renaming them to a second convention, re-keying the metadata and uploading the set to the next platform.
This article covers:
What a connected common data environment is
The reach of a single CDE and where it stops
Why one project ends up with several CDEs
What has to cross a platform boundary
What changes for document control, BIM management and the asset owner
Where a connected common data environment does not help
What is a connected common data environment?
A connected common data environment is an operating arrangement rather than a product category. Two or more CDEs stay under their own owners' control, and an automated layer between them keeps an agreed body of content aligned. Each party retains its licences, its permissions model, its security posture and its internal workflows.
The scope of a connection is defined at the outset. A project decides which folders, which document types and which metadata fields travel, in which direction, and on what trigger. Content outside that definition stays where it is and is not exposed to the other platform.
A one time platform move is a migration and not a sync. Where an organisation is leaving a CDE and taking the record with it, Migration or Sync? Knowing Which One Your Project Actually Needs sets out which of the two a given situation calls for.
What a single CDE does well and where its reach stops
A CDE holds one controlled copy of each information container for the parties inside its licence boundary. It tracks revisions, enforces review and approval, applies suitability codes, and records who did what and when. Teams running one properly spend less time establishing which drawing is current, and can produce the record when a claim arrives.
The authority of a CDE ends at the edge of its own tenancy. It has no visibility of a joint venture partner's platform, no control over a client's mandated environment, and no obligation to understand another vendor's metadata schema. Exchanges across that line fall to people. PlanGrid and FMI, in "Construction Disconnected" (2018), found that construction professionals spend 35 per cent of their working time on non-productive activity, including 5.5 hours a week searching for project data and 4.7 hours resolving conflicts, at a cost of US$177.5 billion a year in wasted labour to the US construction industry.
Why one project ends up with several CDEs
The number of CDEs on a project follows the contract structure. A client mandates its own platform because it will own the asset and wants the record assured on something it controls. A design joint venture standardises on a platform both partners already licence. Each tier one contractor brings the environment its own business runs at enterprise level. Specialist consultants and surveyors arrive with systems tuned to their own deliverables.
The standards accommodate this. Under ISO 19650 at BIM Stage 2, the most commonly implemented level, multiple CDE solutions are permitted on a single project. A single unified collaborative CDE belongs to Stage 3, and few projects have reached that maturity.
Reading the count as non-compliance usually produces a platform mandate. Parties with certified internal document control systems tend to satisfy a mandate by populating the required environment from the one they actually work in. Two environments then exist where the contract intended one. The counting exercise is worked through in our article How Many CDEs Does One Project Have?
What has to cross a platform boundary
Files are the smallest part of what a connection has to carry. A set of documents arriving without its structure, its revision lineage or its approval status gives the receiving team something to sort out before anyone can use it.
Seven things travel together in a working connection. A manual transfer moves some of them and drops the rest.
Metadata carries most of the loss, because schemas rarely agree. One platform holds a single free text status field, another holds a controlled list, and a third splits the same concept across three separate fields. Large public sector projects routinely require in excess of a hundred individual metadata fields per document. Moving those across a boundary needs an explicit field to field map with transformation rules in between. How the platforms differ is set out in ISO 19650 metadata compliance across multiple CDEs.
What changes for document control, BIM management and the asset owner
Document control stops carrying the transfer. A typical large infrastructure project moves 50 to several thousand models and documents a week between CDEs. Automating that movement returns the hours currently spent on downloads and re-keying to register accuracy, transmittal discipline and chasing overdue reviews.
BIM management gets a federation that holds together. Models arriving on the receiving platform with their version history and suitability state intact describe the same design on both sides at the same moment. Clash results and review comments then refer to the same revision wherever they are opened.
The asset owner gets a record it can accept at handover. Bad data cost the global construction industry an estimated US$1.85 trillion in 2020 and caused 14 per cent of all rework globally, worth US$88.69 billion, on Autodesk and FMI Corporation's 2021 survey of more than 3,900 construction professionals. NIST's 2004 cost analysis of inadequate interoperability put the annual cost to the US capital facilities industry at US$15.8 billion, two thirds of it borne by owners and operators during operations and maintenance.
Where a connected common data environment does not help
A connection moves what it is given. Where the source metadata is wrong, the receiving platform gets wrong metadata sooner and in more places. Mapping rules can normalise values and reject a record that fails validation. They cannot supply information that was never captured.
Several other things sit outside the scope of a connection. Ownership of the record is a contract question and stays one. A genuine disagreement about which revision is correct needs a person to settle it. Access rights remain with each platform's owner, so a connection cannot give anyone a permission they were not granted.
Retiring a platform is also outside it. Moving a completed project off a CDE once, with its full history, is handled by CDE Migrate™. The wider set of options for joining platforms, from native connectors to general automation tools to purpose-built platforms, is compared in our CDE integration comparison guide.
Challenges organisations face when connecting CDEs
Schemas that do not line up. Two platforms rarely model status, suitability and discipline the same way, and someone has to decide what each value becomes on the other side.
Service accounts and permissions. A connection needs an account with the right rights on both platforms. A routine access review on either side can remove them without anyone linking the change to the sync.
Agreement on direction. Deciding which platform is authoritative for which content type is a governance decision, and it usually takes longer to settle than the technical configuration takes to build.
Security review on both sides. Each organisation assesses the connection under its own policy, so the timeline is set by the slower of the two reviews. Published control evidence, such as the material on the Security and Trust Portal, shortens that step.
No named owner. Connections built for a particular package often outlive the person who set them up, and a failure goes unnoticed until someone reports missing content.
Is a connected CDE the same as a common data environment?
No. A CDE is one managed environment holding a project's information for the parties inside its licence boundary. A connected common data environment is two or more of those environments kept in agreement automatically, each still owned and administered separately. The connection adds cross-platform behaviour without merging tenancies or changing who controls access on either side.
Does ISO 19650 allow more than one CDE on a project?
Yes. At BIM Stage 2, the level most delivery operates at, multiple CDE solutions are permitted on a single project. A single unified collaborative CDE belongs to Stage 3, which few projects have reached. The requirements apply to the state of the information, its revision, suitability and approval, rather than to the number of platforms holding it.
What is the difference between connecting CDEs and migrating between them?
A migration moves a body of information once, usually when a project changes platform or closes, and the source is then retired. A connection keeps two live environments aligned for as long as both are in use. Some programmes need both, with a migration to establish the baseline and a sync to hold the two sides together afterwards.
CDE Sync™ is Utopia Digital's platform for building a connected common data environment. It sits between platforms as a neutral layer, holds the field level mapping rules, and moves content in both directions on a schedule or on change across 21 supported vendor integrations. Configuration is wizard driven and needs no code, and time to first sync is 15 to 30 minutes. Book a demo to see it run against your own platform mix.