Skip to content

Glossary / S

Server-to-server tracking

What Is Server-to-Server Tracking?

Server-to-server tracking means your website's server sends conversion data straight to an analytics or advertising platform, instead of relying on code running in your visitor's browser.

Most tracking has traditionally worked the other way around. A small piece of JavaScript loads on your page, watches what the visitor does, and reports it. That works fine right up until something stops the script from running, and these days plenty of things do.

Server-to-server tracking skips the browser entirely, since your server already knows an order was placed and can tell the platform itself.

Why It Matters

A good chunk of your visitors never report their conversions through the browser. Ad blockers, tracking protection built into Safari and Firefox, browser extensions, and corporate network filters all stop tracking scripts. A visitor can buy something and the platform never hears about it.

The frustrating part is that you don't get an error. The order goes through, the customer is happy, and the conversion never appears in your reporting. Your ads look worse than they are, and if you're using automated bidding, the platform is making decisions based on incomplete data. Sending the conversion from your server fixes that, because there's nothing in the middle to block.

How It Works Alongside Browser Tracking

Server-to-server tracking usually runs in addition to browser tracking, not instead of it. If both send the same conversion, it would get counted twice, which is why both copies carry a matching identifier. The platform sees two reports with the same ID, works out they're the same conversion, and counts it once. This is called deduplication, and it's what makes running both safe.

Some systems match on a different identifier. Affiliate networks, for example, tend to match on the order reference, since that's the number both copies already share.

What Gets Sent

Server-to-server tracking usually carries more than the browser version can, because your server has access to information the page doesn't.

That typically includes the order total, currency, the products bought, and an order reference. It may also include customer details for matching, like an email address or phone number, which are usually scrambled into an unreadable code before they're sent. Sending those personal details so the platform can match the conversion to a real person is what enhanced conversions are.

One thing the server doesn't automatically have is the click that brought the visitor in. Ad platforms identify which ad earned a conversion using a click ID stored when the visitor arrived. For server-side tracking to attribute correctly, that click ID has to be recorded when the visitor lands and included later when the conversion is sent.

Server-to-Server Tracking on WordPress Sites

On WordPress this is harder than it sounds to build yourself. You need to catch the conversion at the right moment, gather the order details, hold on to the click ID from the original visit, format everything the way each platform expects, hash any customer data correctly, and generate matching IDs so nothing is double counted.

Each platform wants it slightly differently, so doing this for several platforms means several separate implementations.

How Conversion Bridge Helps

Conversion Bridge handles server-to-server tracking for Google Ads, Meta Ads, LinkedIn Ads, Pinterest Ads, Reddit Ads, OpenAI Ads, Snapchat Ads, and Awin, without any code. It stores the click ID when a visitor arrives, sends the conversion from your server when it happens, includes matching identifiers so the browser copy and the server copy are counted once, and formats and hashes customer data to each platform's rules.

You can see which platforms currently support it, along with the rest of what Conversion Bridge connects to, on the platforms page.

Frequently Asked Questions

Does server-to-server tracking replace the pixel?

Usually not, since running both gives you the best coverage. The browser copy arrives right away and the server copy fills in what was blocked, and deduplication stops them overlapping.

Will this double my conversion numbers?

Not when deduplication is set up correctly, which is the normal arrangement. Both copies carry the same identifier and the platform counts one conversion.

Does server-side tracking get around consent requirements?

No, and it shouldn't be thought of that way. If a visitor hasn't agreed to be tracked, that applies whether the data comes from their browser or from your server. Consent is about permission, so it doesn't matter which machine sends the data.

Is it slower for my visitors?

No, the send happens on your server after the conversion, so the visitor isn't waiting on it.