How we work: governance, security, and how you leave.

Escalation

Change requests

Any change to scope, dates or budget is written up with its impact on all three and sent to your sponsor before it is agreed. Nothing is started on a verbal yes. Approved changes appear in the next weekly report with their effect on the plan. On embedded and managed contracts, small changes inside the monthly allowance need no paperwork beyond the ticket.

• Access

  Named accounts only, single sign-on with hardware-key MFA and conditional access policies, least-privilege roles, quarterly access review, and offboarding within one working day. Every credential lives in a company password manager; nothing is shared over chat or email.

• Your code never touches our laptops

  All client code is worked on in virtualised development environments (Coder) behind a VPN, under conditional access policies. Nothing is cloned to an engineer’s device. The environments are destroyed when the engagement ends.

• Devices

  Company laptops enrolled in Microsoft Intune, with full-disk encryption, screen lock, managed updates and endpoint protection. No client data on personal devices.

• Chain of custody for your data

  Every copy of client data is recorded and audited from the day we receive it to the day it is deleted, so we can show where it has been, who touched it, and that nothing insecure remains once the work is done. Deletion is confirmed in writing.

• Changes to your environment as code

  Infrastructure and environment changes are made as code (Terraform) through CI/CD, from the virtualised environments, each with a name and a timestamp. Nobody has direct access to your environment except named DevOps operators, and only if you choose to grant it.

• Development

  Code review on every change, dependency scanning, secrets kept in a vault and never in source, and production changes through a pipeline with an audit trail.

• Incidents

  A written incident process with a named owner, and notification without undue delay and in any case within 72 hours of any incident affecting your data. You also get a client-facing incident portal: report issues, bugs or security concerns and they go straight to the engineering team and are triaged as the top priority.

• Continuity

  Documented runbooks for every managed system, tested restores, and at least two engineers familiar with each client environment.

• Documentation on request

  All of the above is available as written documentation: access and device policies, the environment and change-control model, the chain-of-custody process and the incident process. Ask and it is sent with the proposal.

Insurance, partner programmes, subprocessors and data handling are on trust and assurance, written to be forwarded to a compliance colleague.

• Notice

  Thirty days on every monthly contract: delivery retainers, hosting, and deployment and monitoring. Fixed-price phases end at acceptance. No exit fees, no minimum beyond the stated term.

• Ownership

  You already own the code, the infrastructure definitions, the documentation and the cloud accounts, because they have been in your name from the start. There is nothing to transfer.

• Handover

  Current runbooks and architecture documentation, an inventory of every credential and where it is used, a recorded walkthrough for whoever takes over, and rotation of every secret we held.

• After

  A paid transition period if you want one, at the published day rate, for as long as you need it. Our access is removed on the agreed date and confirmed in writing.

The point of publishing this is that it costs us something to say it, and that is the point. A supplier who cannot describe how you leave is telling you something about how they plan to keep you.

For insurance, partner programmes and data handling see trust and assurance.