Choosing a VPN for business access
Choose a business VPN by what it lets people reach, its device and sign-in requirements, routing, logging, support and what happens when it drops.

A virtual private network creates a protected connection between defined endpoints. For a business, that may mean remote access to office systems or a link between locations.
Start by deciding which connection needs protection and who needs access. A consumer VPN app that hides where your Internet traffic comes from does a different job. It doesn’t give staff controlled access to business systems.
Define the resource and the permitted users
List the applications or network services that remote staff need. Decide which users and devices may reach each one.
Ask whether the product opens the whole office network to a connected device or only particular applications. Per-application access (often sold as zero trust network access) limits what a stolen login can reach. Include suppliers and administrators, whose needs may differ from other staff.
The Cyber Centre’s VPN guidance describes selection and configuration considerations. Review the design with the person responsible for your network.
Review endpoint and identity requirements
Check the supported operating systems, client software and device-management requirements. Include the process for a lost device or a departing user.
Require multi-factor authentication for every VPN sign-in. Confirm who approves new access and how a locked-out user gets back in. Try the sign-in policy on a test account first.
A protected tunnel doesn’t remove malware from a device or prevent a user from responding to a fraudulent request. You still need endpoint protection on the device and a habit of checking unusual requests.
Understand routing and information handling
Ask which traffic uses the tunnel and which goes directly to the Internet. Have IT document the reason for that routing and the controls applied to each path.
Check DNS handling, logging, log access and retention. Find out what the provider can see, where it keeps logs and which country’s law applies.
A no-logging promise doesn’t make your business activity anonymous. Applications and destination services can still hold their own records.
Test performance and interruption behaviour
Try the applications from the locations and connections staff use. Include a weak connection, a large transfer and a meeting if those tasks are part of the requirement.
Check what happens when the tunnel disconnects or the computer changes networks. Decide which work must stop, which can continue another way, and what that way is.
If the client has a kill switch, which blocks Internet traffic while the tunnel is down, test it. It can also stop work staff expect to carry on, so tell them what to do and who to call.
Verify maintenance and support
Identify who patches the client and gateway, monitors access and investigates failed sign-ins. The Cyber Centre’s edge-device guidance covers the security responsibilities of equipment exposed at a network boundary.
Check vendor support dates, licensing, service contacts and how access is revoked. Rule out anything that still relies on PPTP, which has been insecure for years. Plan for an outage when your usual administrator is away.
Compare the full cost of each option that passed your tests. Keep the configuration notes and test results, so a later change can be checked against the same access rules.


