Four service lines, delivered by engineers who have moved an estate like yours before. They work alongside your people rather than in place of them, and the engagement is designed to end.
Most customers take implementation first and add one of the others within a year. They can be bought separately and are priced separately.
Getting the first products into production, configured for how your organisation actually works rather than how the demonstration tenant does.
An implementation begins with two weeks of discovery run against your live systems, not against a questionnaire. We come out of it with a configuration design, a data model, a list of the integrations that have to exist on day one, and an explicit list of the things we are choosing not to do yet.
From there the work is done in fortnightly increments in your own tenant, with your people in the room. Nothing is built in a vacuum and handed over: the team that will run the system afterwards configures a meaningful part of it during the project, which is the only handover method that survives contact with the first incident.
Moving twenty years of records off a system nobody has fully understood since the person who built it left.
Migration is the part of a platform change most likely to fail, and it fails for the same reasons every time: the source data is worse than anyone believed, the reconciliation was left until the end, and the cutover was rehearsed once at a tenth of production volume.
We invert all three. Profiling comes first and produces a written statement of what cannot be mapped, before any commitment is made. Reconciliation is designed alongside the mapping rather than after it. And the cutover is rehearsed at full volume at least twice, with the rollback timed as carefully as the migration itself.
We run the estate for you: releases, monitoring, tenant administration and the on-call rota, against a service level you can hold us to.
Managed operations started because customers kept buying the platform and then failing to staff it. A system nobody owns degrades quietly, permissions drift, integrations silently stop, and nobody notices until an auditor does.
Under a managed engagement a named operations team owns your tenant configuration, applies platform releases in your change window, monitors the integrations you depend on, and carries the pager. You keep every administrative right; we simply do the work. The engagement is reviewed quarterly against a published set of measures, and it can be ended with ninety days' notice without any technical unwinding.
Teaching your administrators and your builders to need us less, which is the only sensible definition of a successful engagement.
Training is delivered by the engineers who build and implement the platform rather than by a separate faculty, which means the examples are real and the answers to awkward questions are accurate.
There are three tracks. Administration covers identity, entitlements, tenant configuration and the audit trail. Building covers Forge, the integration runtime and the event model, for organisations that intend to write their own applications. Operations covers monitoring, release management and incident handling, and is a prerequisite for taking a managed engagement back in house.
A representative sixteen-week implementation. The shape holds for longer programmes; the phases stretch rather than multiply.
Two weeks inside your live systems with the people who use them. Produces a configuration design, a data assessment and an explicit scope boundary. If the answer at the end is that you should not do this project, we say so and you owe us for the discovery only.
The configuration design, entitlement model, integration list and migration approach are reviewed and signed by named owners on your side. Fixed-price implementations are priced from this document, not from the sales conversation.
Fortnightly increments in your tenant, demonstrated to the same group each time. Your administrators configure a defined share of the system themselves, supported by our engineers rather than replaced by them.
Full-volume migration rehearsals with timed rollbacks, user acceptance testing against real scenarios, and a cutover runbook rehearsed end to end before the weekend it matters.
The same engineers stay on for four weeks after go-live, and through the first period close where finance is involved. Issues raised in hypercare are fixed rather than logged.
Either your team takes the system, with the operations training completed and the runbooks written, or a named managed operations team takes it. Both endings are planned from the start; neither is a default.
Consultancies rotate their good people onto the next sale. We staff an engagement with a lead engineer, one or two implementation engineers and — where data is moving — a migration consultant, and those names are in the statement of work.
Every engineer on the delivery team has shipped production configuration on the platform and carries the same certification we ask of your administrators. Around a third of them joined from customers who had been through an implementation with us.
If we have to change someone, the replacement overlaps for two weeks at our cost. That is written into the contract rather than promised in a meeting.
Support is separate from professional services and is priced separately. Every subscription includes the standard tier; the other two are optional and can be taken per product rather than across the estate.
Included with every subscription
Severity one response
Next working day
Priced as a percentage of subscription
Severity one response
Two hours, 24 × 5
Available with managed operations
Severity one response
Thirty minutes, 24 × 7
It produces a configuration design and an honest assessment of your data. If that assessment says the project should not proceed, you owe us for the discovery and nothing else.
One hundred and seven business applications, one identity layer, one audit trail. Buy what you need; the rest is waiting when you do.
CEDX Systems LimitedSixth Floor40 BishopsgateLondon EC2N 4AJUnited Kingdom© 2026 CEDX Systems Limited. Registered in England and Wales, company number 07648321. VAT registration GB 184 7752 09.
cedxsystems.com
