
Site to Site VPN: How It Securely Connects Offices, Branches and Cloud Networks
A site to site vpn connects whole networks, not just one laptop, so staff at different offices can use shared files, printers, internal apps and cloud resources as if they were on the same private network. Traffic travels through an encrypted tunnel across the public internet, which protects it from snooping while avoiding the cost and delay of a private leased line for every site. This guide explains how it works, how it differs from a remote-access VPN, how to set one up safely, and the mistakes that most often break new deployments.
What Is a Site to Site VPN?
A site to site vpn is a permanent or always-on encrypted connection between two or more networks. Each site has a VPN gateway, usually a firewall or router, and the gateways authenticate each other before exchanging traffic. Once the tunnel is up, devices at one site can reach approved resources at another site without installing VPN software on every computer. The connection is transparent to most users: they open a shared drive or internal application, and routing sends that traffic through the tunnel automatically.
How a Site to Site VPN Works
When a device sends data to another site, its local gateway checks the destination, encrypts the packet and wraps it for transit across the internet. The receiving gateway verifies the sender, decrypts the packet and delivers it to the local network. IPsec is the most common protocol family for this job: IKE negotiates the security association and keys in phase 1 and phase 2, while ESP encrypts and authenticates the data itself. Routing rules, often static routes or a dynamic protocol such as OSPF or BGP, decide which traffic enters the tunnel and which traffic goes directly to the internet.
For reliability, many designs use dead peer detection to spot a failed peer, redundant tunnels or a second internet connection, and monitoring for tunnel status, latency and packet loss. Without those checks, a tunnel can fail silently and leave a branch office cut off from central systems until someone notices.
Types of Site to Site VPN
Not every site-to-site design is the same. The right choice depends on whether the sites belong to one company, a partner, or a cloud provider.
Intranet-Based VPN
An intranet-based VPN connects LANs owned by the same organization into one WAN. It is the classic headquarters-to-branch design and works well for shared ERP, file and intranet systems.
Extranet-Based VPN
An extranet-based VPN connects networks belonging to different organizations, but only for agreed resources. It is useful for suppliers or joint projects, provided access is tightly segmented and reviewed regularly.
Internet-Based and MPLS-Based VPNs
Internet-based tunnels run over the public internet and are usually the lowest-cost option. MPLS-based VPNs run over a provider-managed private network and can offer more predictable performance, but often at higher cost and with less direct control.
Cloud Site to Site VPN
Cloud VPNs connect an office network to a virtual private cloud in AWS, Azure or Google Cloud, or connect VPCs in different regions. They are essential for hybrid cloud work, but cloud routing, address ranges and firewall rules must be planned before the tunnel is built.
Benefits of a Site to Site VPN
The main benefits are stronger protection for inter-office traffic, simpler access to shared resources, lower cost than many private lines, and easier expansion: adding a site usually means configuring one more gateway rather than rebuilding the network. Central security policies can also be applied consistently, and staff do not need to remember to start a VPN client.
Limitations to Plan For
Site-to-site VPNs are not a complete security solution. They can be complex to scale in a full mesh, add latency, and create broad trust between networks if firewall rules are too permissive. They also do not protect a compromised device inside a trusted site, so endpoint security and segmentation still matter.
Site-to-Site VPN vs Remote-Access VPN
The key difference is scope. A site-to-site VPN connects networks; a remote-access VPN connects one person or device to a network. Use the table to compare them quickly.
| Feature | Site-to-Site VPN | Remote-Access VPN |
|---|---|---|
| What it connects | Entire networks | Individual users or devices |
| Who uses it | All staff at a site, automatically | Remote workers, one device at a time |
| Setup | Gateways at each site | VPN client on each device |
| Availability | Usually always on | Usually on demand |
| Best for | Offices, branches and cloud VPCs | Working from home or travel |
| Security model | Network-to-network trust, needs segmentation | User-to-network access, needs MFA and device checks |
In practice, many organizations need both: site-to-site tunnels for offices and cloud networks, and remote-access VPN or Zero-Trust Network Access (ZTNA) for people working outside those sites.
How to Set Up a Site to Site VPN: Overview
Exact screens vary by vendor, but the sequence below works for most firewall and cloud VPN deployments.
- Plan non-overlapping private address ranges for every site and cloud VPC.
- Choose IPsec with IKEv2, strong encryption and a matching proposal on both gateways.
- Configure each gateway with its public address, tunnel interface and local networks.
- Exchange authentication securely, preferably with certificates rather than a weak shared secret.
- Define which traffic enters the tunnel and configure static or dynamic routing.
- Set firewall policies so only approved services can cross between sites.
- Test traffic in both directions, check failover, and turn on tunnel monitoring and logging.
Security Best Practices
Use AES-256 encryption, strong IKEv2 settings, and certificates or long random pre-shared keys stored securely, never short shared passwords. Limit tunnel traffic with firewall rules, segment sensitive systems, and log tunnel events centrally. Rotate keys, patch gateways promptly, enable dead peer detection and redundancy, and review who can reach what across the tunnel at least quarterly.
Common Mistakes That Break a Site to Site VPN
Most failures are planning failures, not protocol failures. Check these before go-live:
- Overlapping private IP ranges: two sites using the same subnet cannot route cleanly to each other.
- Mismatched IKE or IPsec proposals: encryption, hashing or lifetime settings must agree on both ends.
- Firewall rules that allow all traffic: a tunnel should not become an open bridge between networks.
- A single tunnel with no failover: one internet outage or gateway failure should not stop the business.
- Weak pre-shared keys: short keys, or keys shared by email or chat, undermine the whole tunnel.
- No monitoring: without alerts, outages are discovered by users instead of by IT.
A short pre-launch checklist — test from both directions, confirm DNS and routing, simulate a gateway failure, and document the design — prevents most of these problems.
When to Choose Each Type
Choose site-to-site when whole offices, warehouses or cloud VPCs must share resources continuously. Choose remote-access when individual people need secure access from home, hotels or mobile networks. Choose both when offices are connected and staff also travel. If access must be limited to specific applications rather than whole networks, consider SASE or ZTNA alongside, or instead of, a broad VPN.
Frequently Asked Questions
What is a site to site vpn in simple terms?
It is an encrypted tunnel between two or more whole networks, such as a head office and a branch, so approved devices at each site can share resources securely over the internet.
Is a site to site vpn secure?
Yes, when it uses strong IPsec encryption, secure authentication, patched gateways and strict firewall rules. Security weakens quickly if keys are weak, rules are too broad, or gateways are left unpatched.
Do users need VPN software on every device?
No. The gateways handle the tunnel, so most office devices need no VPN client. Individual remote workers outside those sites still need remote-access VPN, ZTNA or a similar client-based method.
How is it different from a remote-access VPN?
A site-to-site VPN connects networks to each other and is usually always on, while a remote-access VPN connects one user or device to a network, usually on demand.
What causes a site-to-site VPN to drop?
Common causes include internet outages, mismatched IKE or IPsec settings, expired keys or certificates, overlapping IP ranges, firewall changes, and gateway reboots without a redundant tunnel.
A well-planned site to site vpn gives offices and cloud networks a private, encrypted path to shared resources without forcing every user to manage a client. Plan addressing, lock down firewall rules, test failover, and monitor the tunnels, and the result is a connection that is both usable and defensible.
Before you launch, document each gateway, tunnel and firewall rule, and keep a tested rollback plan. That record makes troubleshooting faster, onboarding easier, and future audits far simpler for your whole team.


