System Architecture & Design Flashcards

What is the primary purpose of a data layer in a web analytics implementation?

Answer: To provide a structured, page-level JavaScript object that tag managers read instead of scraping the DOM

A data layer decouples tracking from page markup so tags read stable, structured values rather than fragile DOM elements.

Which architecture sends events from the browser to your own server endpoint before forwarding them to analytics vendors?

Answer: Server-side tagging

Server-side tagging routes hits through a first-party server container that then relays data to vendors.

A key benefit of serving the tracking endpoint from a first-party subdomain is that it:

Answer: Can extend cookie lifetime and reduce blocking by browsers and ad blockers

First-party endpoints are less likely to be blocked and can set longer-lived server-set cookies, though consent is still required.

Why is a tag management system typically loaded asynchronously?

Answer: So it does not block page rendering

Asynchronous loading prevents the tag container from delaying page rendering and hurting user experience.

In a typical web analytics data pipeline, what is the correct order of stages?

Answer: Collection → Processing → Storage → Reporting

Data is collected, processed (filtered/enriched), stored, and finally surfaced in reports.

Which problem is most likely when two different tags both fire a 'purchase' event for the same transaction?

Answer: Duplicate revenue and conversion inflation

Duplicate firing double-counts transactions, inflating revenue and conversions unless de-duplicated by transaction ID.

What is the advantage of using a unique transaction ID in ecommerce tracking?

Answer: It allows de-duplication of repeated purchase hits

Analytics platforms can discard repeat hits carrying a transaction ID already recorded.