Somebody has to own it before it breaks.

The gap between working and operated

The network, the server, the backups, the logins and the equipment in the corner all work, so they receive no attention. Then something stops on a Tuesday morning and the answer is a phone call to whoever set it up, who may or may not remember it and may or may not still be available.

Splide takes that estate over as an ongoing operation rather than a project or a support queue.

The value is not only in answering when something breaks. It is in knowing what exists, how it is connected, what changed, what is exposed and what is likely to fail next.

What we actually run

This is not a package menu. It is the shape of an estate under management today for an engineering consultancy and its fabrication plant.

Networks

Switching, wireless, firewalls, remote access and links between an office and a plant floor that are not the same environment and should not be managed as though they are.

Windows servers and identity

Domain controllers, file services, group policy, accounts and permissions. The invisible work that decides who can open what, and what remains accessible after somebody leaves.

Backups that are tested

Jobs that run and restores that are actually attempted. An untested backup is a belief, not a backup.

Monitoring that reaches an owner

Alerts go to the person expected to act on them. That is the difference between a disk filling quietly for three weeks and a disk being dealt with before the operation stops.

Equipment that is not office IT

A fabrication floor runs industrial controllers and CNC equipment on a network that cannot be treated like a row of desks. Keeping that environment working is part of the operation, not an awkward exclusion discovered after the contract starts.

One accountable place

When something breaks, there is one place to call. The person answering already knows how the estate fits together because Splide runs it continuously.

The system remembers the estate

Conventional IT support is organised around tickets. Tickets remember individual incidents badly and relationships between them hardly at all.

Splide built Nova to operate an estate as a connected system: discovery, inventory, encrypted credentials, diagnostics and integrations across firewalls, virtualisation, networking and Active Directory.

Agents can work with that context and with the history held in Adja. They help connect a new symptom to an old change, keep an investigation from restarting in every session and preserve what was learned after the incident closes.

AI changes what one operator can hold in view. It does not move accountability. Splide remains the operator responsible for deciding and acting.

Nova is not a product and is not for sale. It is part of the machinery behind the service.

What this is not

It is not a support contract that begins when somebody reports a fault. A provider paid only when things break has the wrong incentive.

It is not automatically a rebuild. Most installed equipment is serviceable. The common problem is that nothing has owned the whole estate for a while.

It is not a menu of websites, apps and once-off installations. Infrastructure operation is one of Splide's three ongoing responsibilities, and it is treated accordingly.

Pricing

The fee is monthly and agreed against the estate that actually exists after an audit, not against an arbitrary package tier.

The fee buys operations. It never buys permission. Credentials, licences, equipment and systems remain the client's. Ending the service ends Splide's responsibility for running them; it does not remove access to them.

Starting

The first step is an audit rather than a proposal.

What is running? What is exposed? What has no tested backup? Which credentials are unknown? Where does one failure stop the whole business? What is going to fail first?

The client keeps the audit whether or not Splide takes over the estate.