Many site owners treat video calling as a plugin problem. In reality, it is a bandwidth and moderation problem. The browser side was solved years ago. Everything behind it costs money every single month.
So, this guide breaks the real costs down, line by line. Read it before you quote a client a number for a live video feature.
Why Demos Are Cheap and Production Is Not
WebRTC ships inside every modern browser. Therefore you pay nothing to use the technology itself. Your visitors need no plugin and no install.
As a result, prototypes look almost free. A developer builds a working two-person call in a weekend. The client sees it running and assumes the feature is nearly finished.
However, that demo connects two browsers directly. Production traffic does not behave that way, and the gap between the two is where the budget disappears.
The Four Cost Centers
| Cost centre | What drives it | Rough order of magnitude |
|---|---|---|
| TURN relay bandwidth | 15–20% of calls that cannot connect directly | ~1 GB per hour of relayed call |
| Signalling server | Uptime, not capacity | Low, but must never go down |
| Moderation | Minutes of live video, not messages | The largest line at scale |
| Age assurance | Signups, plus lost conversion | Small per check, large in drop-off |
Each one is explained below.
Cost 1: TURN Relay Servers
Roughly 15% to 20% of connections cannot link peer to peer. Corporate firewalls and strict mobile carriers block them. Those sessions need a TURN relay instead.
A relay carries the full audio and video stream. Therefore you pay for every gigabyte that passes through it.
In short, budget realistically. A single one-to-one call at decent quality uses about 1 GB per hour across both sides. Consequently, a thousand relayed hours per month becomes a visible line in your hosting bill rather than a rounding error.
Self-Hosted or Managed?
First, you can run TURN yourself with coturn, or buy it from a provider.
Self-hosting lowers your per-gigabyte cost significantly. However, it adds servers you must monitor, patch and scale. For a small team, a managed provider is usually cheaper once you price your own time honestly.
Cost 2: Signalling
Before two browsers connect, they exchange connection details. That exchange runs through your signalling server.
Signalling uses very little bandwidth. Therefore the direct cost stays low. However, availability matters far more here than capacity does. If signalling goes down, nobody can start a call at all, even though every other part of your stack is healthy.
For this reason, most teams run this as a small WebSocket service. Also, keep it separate from your main WordPress host so a traffic spike on the site cannot take calls offline.
Cost 3: Moderation
This is where projects fail, and it is almost always missing from the original quote.
Text can be scanned after publication. Live video cannot. If strangers will talk to each other on your site, you need moderation during the session itself.
Consequently you need automated frame analysis, human reviewers, or both. Automated screening costs money per session minute. Human review costs considerably more. Moreover, neither works without a reporting flow, a blocking system and somebody who actually reads the reports.
Why Retrofitting Is Expensive
Above all, plan this before launch. Adding moderation to a live product afterwards means rebuilding session handling, storage and your admin tools at the same time.
In addition, the cost scales with usage rather than with headcount. That makes it behave unlike almost every other line in a typical web project budget.
Cost 4: Compliance Changed the Maths
Age assurance rules now apply in the UK and in several US states. Therefore a site that connects strangers may need verified age checks before the first session begins.
Self-declaration no longer counts in the UK. Instead, regulators expect methods such as facial age estimation, open banking checks or ID matching.
Admittedly, each check costs a small amount per user. However, the real cost is conversion. Every verification step removes a share of your signups, and a product whose value depends on having enough people online suffers twice from that drop.
So Should You Build It At All?
To decide, ask one question first. Is live video your product, or is it a feature of something else?
If it is your product, build it. Budget for relay bandwidth, moderation and compliance from day one, and treat all three as ongoing operating costs.
If it is a feature, then consider two cheaper routes instead.
Route A: A Managed API Provider
Several companies handle relay, scaling and sometimes moderation for you. You pay per minute and skip most of the infrastructure work.
In particular, this suits teams who need calling inside their own product and do not want to run servers. The trade-off is margin: at high volume, per-minute pricing becomes the dominant cost.
Route B: Link Out Instead of Building
For example, a community site or a membership platform can point members toward an existing video chat service rather than hosting sessions itself.
As a result, you keep the social benefit for your members and avoid relay bills, moderation staffing and the entire compliance burden. That option sounds unambitious. However, it is the correct answer far more often than developers like to admit, particularly when calling is a nice-to-have rather than the reason people visit.
A Checklist Before You Quote
Run through these points before giving any client a figure:
- Estimate monthly relayed minutes, not total minutes.
- Confirm in writing who pays for bandwidth overage.
- Decide between automated moderation, human moderation or both.
- Check whether age assurance applies in your client’s markets.
- Add a line for reporting and blocking tools.
- Include ongoing maintenance, not just the initial build.
- Agree what happens if usage grows ten times faster than forecast.
Common Mistakes Developers Make
Meanwhile, a few patterns show up repeatedly on projects that go over budget.
Quoting from the prototype. The weekend demo used direct connections and no moderation. It is not a sample of production.
Forgetting mobile networks. Carrier NAT pushes more sessions onto relay than office Wi-Fi testing suggests.
Treating moderation as a phase two item. It is a launch requirement in most jurisdictions now, not a later improvement.
Ignoring the conversion cost of verification. Teams model the per-check fee and never model the signups they lose at that step.
Final Thoughts
To sum up, video chat is not expensive to prototype. It is expensive to run safely at scale.
Overall, the browser gives you the hard engineering for free. In exchange, you inherit bandwidth, moderation and compliance costs that grow in direct proportion to how successful the feature becomes.
Therefore, treat live video as an ongoing service rather than a one-off build. Price it that way from the start, and the project will survive its first genuinely busy month.
