Server-Side Conversion Tracking
In some business scenarios, important outcome signals do not originate in the user’s browser.
Examples include:
- payment confirmations processed by backend systems
- subscription status changes
- CRM or order management events
- webhook-driven fulfilment updates
In these cases, conversions should be recorded server-side to ensure accuracy and resilience.
When to Use
Server-side tracking is recommended when:
- conversion logic is finalised only on backend systems
- browser tracking could be blocked by ad-blockers or consent timing
- sensitive business events should not be exposed in client code
- multi-system orchestration determines the real outcome
Example
import { createNodeMark } from '@crelora/mark/node'
const mark = createNodeMark({
key: process.env.MARK_SECRET_KEY,
site_id: process.env.MARK_SITE_ID,
site_host: process.env.MARK_SITE_HOST
})
mark.conversion('Subscription Activated', {
order_id: 'sub_456',
value: 4900,
currency: 'eur'
})
In this setup:
- conversion signals are triggered directly from trusted backend logic
- secret keys are used instead of publishable browser keys
- outcome data is less dependent on client-side execution conditions
Hybrid Conversion Tracking
For many modern growth environments, the most accurate approach is a hybrid model:
- the browser records behavioural intent signals
- the backend confirms the final business outcome
- both signals are linked through a shared visitor identifier
Example:
Browser side:
const visitorId = Mark.getVisitorId()
fetch('/api/checkout', {
method: 'POST',
body: JSON.stringify({
visitor_id: visitorId,
cart_value: 12900
})
})
Server side:
mark.conversion('Purchase Confirmed', {
visitor_id: 'vis_abc',
order_id: 'ord_789',
value: 12900
})
This model allows OneLence to:
- connect marketing attribution with verified business outcomes
- reduce signal loss caused by browser limitations
- support advanced revenue analytics and optimisation models
In practice, hybrid conversion tracking provides the highest level of attribution reliability.
