Your VPN Is Twenty Years Old Too: Redesigning Remote Access for the Way Your SME Actually Works Now

  • 2 weeks ago
  • 0

A VPN can still be doing its job while the way your organisation works has changed around it. Staff may now use cloud applications, private business systems, phones and personal devices from a variety of locations. Yet the remote-access design may still assume that once someone connects to the network, they can be trusted to move around it.

That assumption is the real legacy. The issue is not simply how old the VPN appliance is, but whether the access it grants still matches the business need.

Replacing it with a newer box may improve reliability or address a support deadline. It will not, on its own, answer the strategic question: which people and devices need access to which services, under what conditions?

A VPN is a connection, not a trust decision

A VPN encrypts traffic between a device and a network. That can be valuable, particularly where a business needs to protect traffic in transit or support a system that has no other practical remote-access route. The National Cyber Security Centre (NCSC) recognises those uses, while noting that VPNs are complex to configure and maintain, and that a failure can disrupt an organisation.

But an encrypted tunnel does not automatically mean that the user should be able to reach every system behind it. In a broad, traditional setup, a successful connection can grant more network reach than a person needs. If an account or device is compromised, that excess reach can increase the impact.

The NCSC’s VPN guidance helps organisations assess when and how to use VPNs. Its Zero Trust guidance describes a different direction: do not treat a network location as proof of trust; assess access requests and grant only what is needed.

That does not mean every SME should switch off its VPN. It means “connected” should no longer be shorthand for “trusted”.

Design access around the application

For many modern services, the requirement is not “put this person on the company network”. It is “let this person use this particular application”. Zero Trust Network Access (ZTNA) is intended to support that more specific model, making access decisions before a connection is granted and using relevant information about the user, device and service.

The NCSC’s ZTNA implementation guidance emphasises that encrypted transport is important but is not a substitute for access policy. A robust design should consider whether the device is known and in a suitable security state, which application is being requested, and whether the request should be allowed at that moment.

A laptop and phone linked to distinct service nodes, representing access limited to specific applications

In practice, that could mean staff access cloud services directly through their organisational identity and application controls, while private applications are made available through a tightly scoped gateway or connector. Access should be limited to the systems required for a role, rather than opening a general route into the internal network.

This is an architecture decision, not a product-shopping exercise. The right design depends on the services in use, the sensitivity of the information, the devices involved and the business consequences of losing access. A consultancy-first review can map those requirements before deciding which technologies, if any, need to change.

Bring personal devices into the design, not through the back door

BYOD makes remote access more complicated because the organisation does not fully control a personally owned device. Staff may reasonably expect privacy, while the business remains responsible for protecting its own data and systems.

The NCSC’s BYOD guidance highlights the importance of deciding what the organisation wants BYOD to achieve, understanding the risks and choosing suitable controls. The key design question is not simply whether a personal device can connect. It is what information and capability that device should be allowed to reach.

A proportionate SME policy might allow limited access to selected services from personal devices, while reserving sensitive data and administrative functions for managed equipment. Depending on the service and risk, controls might include application-level access, separating business data from personal data, restricting downloads, or providing a managed workspace. The organisation should also be clear about how corporate data can be removed without wiping personal content.

Per-app VPN may help in some BYOD situations, but it is not automatically equivalent to ZTNA. The NCSC notes that the protected application set must be defined, and that traffic or data may escape the intended controls through other applications or configuration gaps. Treat it as a design choice with explicit limits, not a security label.

Personal devices beside a contained work-service area, illustrating clear boundaries for BYOD access

Redesign in stages, without breaking the working day

For most SMEs, remote-access redesign is safer as a sequence of controlled changes than as a one-night cutover.

  1. Understand what the VPN actually supports. Map the applications, network routes, user groups, devices and business processes that depend on it. Identify routes that are unused, overly broad or retained for a specific legacy requirement. Ask staff and service owners what they need to do remotely, rather than assuming the current configuration reflects today’s work.

  2. Reduce unnecessary reach first. Review access groups and narrow routes so that users can reach only the services required for their roles. Separate ordinary staff access from administrator access, and set a higher bar for sensitive systems. This can reduce exposure while the existing VPN remains in service.

  3. Move suitable applications to more specific access. Pilot identity-aware, application-level access for services that can support it. Begin with a manageable group and a small number of applications. Test real user journeys, access decisions, support processes and failure recovery before expanding the change.

  4. Contain systems that cannot move yet. Older software and specialist workloads may not support modern access methods. Keep them behind a restricted, monitored route, proxy or segmented network rather than giving them a broad exception. For example, organisations relying on IBM i management may have specific operational dependencies to assess as part of the access design.

  5. Retire access paths only when their purpose has gone. Confirm that each remaining application has an appropriate secure route, that users can still perform essential work and that support teams can diagnose access problems. Remove obsolete routes and client profiles in measured waves, with a tested fallback for important services.

Distinct service zones connected by controlled pathways, illustrating a staged transition from legacy remote access

The NCSC’s ZTNA guidance also stresses limiting application exposure, monitoring access decisions and designing for resilience. These are practical requirements, not finishing touches. A new access service that blocks legitimate work without a recovery process can quickly lose user confidence; a service with poor visibility can leave support teams guessing.

Make the business case measurable

Remote access redesign should be judged by business outcomes, not by the number of VPN licences removed. Useful measures include how many services are reachable from a standard user connection, how quickly unnecessary access can be removed, how well the organisation can identify unusual activity and how reliably staff can reach essential applications.

A Cloud Solutions review can help establish which services suit direct, identity-led access. Network Solutions can support segmentation and controlled routes for private systems. Disaster Recovery planning should consider how access will be restored if a critical service or access component fails. The operational work can then be supported through Managed IT Services and Helpdesk Support, as execution services beneath the broader strategy.

There may also be a cost-control opportunity. Reviewing unused remote-access capacity, overlapping services and telecoms arrangements can form part of Technology Expense Management. Where remote staff rely on business calling, Voice and Connectivity should be considered alongside application access, so that voice requirements do not become an unexamined reason to preserve an overly broad tunnel.

For businesses coordinating office teams and field-based property operations, access needs can differ sharply between roles. Evestaff’s group of services can be explored through its gateway website.

Strategy first, technology second

A VPN is not obsolete merely because it is old, and Zero Trust is not a switch that makes every risk disappear. The useful question is whether the current access model grants the right people, on appropriate devices, access to the right services, and whether that decision can be reviewed and supported.

A Fractal IT Director approach starts with that business-level design: understand workflows, risk and cost; establish the target access model; then plan the technology and operating changes. Specialist channel partners can provide specific implementation capacity where needed, while traditional support services remain bolt-ons that help execute and maintain the strategy.

The destination may be ZTNA for modern applications, a tightly controlled VPN for systems that still require it, or a mix of both. The important step is to make that mix deliberate, measurable and aligned with how the SME actually works now.

SEO keywords:

Cloud Solutions, Disaster Recovery, Dynamics 365 Jump Start, Helpdesk Support, IBM i Management, Managed IT Services, Network Solutions, Technology Expense Management, Voice and Connectivity

Join The Discussion