Comprehensive knowledge of analytics platforms, implementation of tracking, reporting infrastructure, and dashboard design to support marketing, product, and content decisions. Candidates should be able to describe tool selection and configuration for platforms such as Google Analytics Four, Adobe Analytics, Mixpanel, Amplitude, Tableau, and Looker, including the trade offs between vendor solutions, native platform analytics, and custom instrumentation. Core implementation topics include defining measurement plans and event schemas, event instrumentation across web and mobile, tagging strategy and data layer design, Urchin Tracking Module parameter handling and cross domain attribution, conversion measurement, and attribution model design. Analysis and reporting topics include funnel analysis, cohort analysis, retention and segmentation, key performance indicator definition, scheduled reporting and automated reporting pipelines, alerting for data anomalies, and translating raw metrics into stakeholder ready dashboards and narrative visualizations. Integration and governance topics include data quality checks and validation, data governance and ownership, exporting and integrating analytics with data warehouses and business intelligence pipelines, and monitoring instrumentation coverage and regression. The scope also covers channel specific analytics such as search engine optimization tools, social media native analytics, and email marketing metrics including delivery rates, open rates, and click through rates. For junior candidates, demonstration of fluency with one or two tools and basic measurement concepts is sufficient; for senior candidates, expect discussion of architecture, pipeline automation, governance, cross functional collaboration, and how analytics drive experiments and business decisions.
EasyTechnical
51 practiced
How would you explain funnel analysis to a non-technical stakeholder? Describe what a funnel shows, how drop-off is interpreted, and when a funnel is more useful than a simple conversion rate.
Sample Answer
A funnel shows how many users move through a defined sequence of steps, such as landing page -> signup -> verification -> purchase.**How to explain drop-off**- Each step shows how many users made it that far- The gap between steps is drop-off- Drop-off tells you where users lose interest, hit friction, or encounter errors**Why it’s useful**A simple conversion rate tells you the final outcome, but a funnel shows where the problem happens. For example, if overall signup conversion drops, the funnel can reveal whether the issue is the form, email verification, or payment step.**When I’d use it**- To diagnose UX friction- To compare performance across devices, channels, or versions- To prioritize fixes based on the biggest loss pointAs a data engineer, I’d make sure the funnel events are consistently defined and ordered so stakeholders aren’t comparing apples to oranges. Good funnel data helps teams focus on the exact step that needs improvement.
HardTechnical
52 practiced
Users often start on one device, continue on another, and later log in. How would you design identity resolution for analytics so product and marketing reports can connect anonymous and authenticated behavior without over-merging unrelated users?
Sample Answer
I’d use a hybrid identity graph with strict merge rules, because over-merging is usually worse than missing a few links.**Signals I’d use**- Deterministic: login, account creation, email hash, authenticated session.- Probabilistic only if policy allows, and never as the sole merge criterion.**Modeling approach**- Keep anonymous_id, device_id, and user_id as separate entities.- Create a bridge table that maps identifiers to a canonical person or account ID.- Preserve event history so we can re-run identity logic if rules change.**Merge rules**- Merge only when confidence is high and the relationship is explainable.- Add time bounds so an old anonymous cookie doesn’t get linked forever.- Prevent cross-account merges when a device is shared.**Reporting impact**- Product analytics can use stitched user journeys.- Marketing attribution can connect pre-login and post-login touches.- Dashboards should show merge coverage and confidence so users know how complete the view is.The key is to favor precision over recall, because false merges distort both retention and attribution.
EasyTechnical
56 practiced
You're explaining product analytics to a new product manager. What is the difference between a page view, session, user, and event in GA4 or Amplitude, and why do those definitions matter when comparing reports across teams?
Sample Answer
In GA4 or Amplitude, these terms mean different things, and the definitions affect every dashboard downstream.**Page view**: a page loaded or viewed. On a content site, this is often the basic content consumption unit.**Session**: a period of activity from a user, usually grouped by inactivity rules. One user can have many sessions.**User**: a unique person or device identified by an ID or cookie. User counts answer "how many distinct people used this?"**Event**: any tracked action, such as click_button, signup, purchase, or video_play. Modern analytics tools are event-based, so this is the most flexible unit.**Why it matters**- Teams may compare different scopes by accident: one report may use users, another sessions.- Session definitions differ across tools, so cross-platform comparisons can be misleading.- A page view-heavy report can look healthy even when events show low engagement.As a data engineer, I’d make sure metric definitions are documented in a semantic layer or data dictionary so analysts and PMs are comparing the same thing.
MediumTechnical
50 practiced
Write a SQL query to calculate 7-day retention by signup cohort and acquisition channel from an events table with user_id, event_name, event_time, signup_time, and channel. Explain how you would handle late-arriving events and duplicate rows.
Sample Answer
I’d deduplicate first, then build the cohort retention query off the first signup date and the acquisition channel.
sql
WITH deduped AS (
SELECT *
FROM (
SELECT
user_id,
event_name,
event_time,
signup_time,
channel,
ROW_NUMBER() OVER (
PARTITION BY user_id, event_name, event_time
ORDER BY event_time
) AS rn
FROM events
) t
WHERE rn = 1
),
cohorts AS (
SELECT
user_id,
DATE_TRUNC('day', signup_time) AS signup_cohort,
channel,
signup_time
FROM deduped
WHERE event_name = 'signup'
),
retained AS (
SELECT DISTINCT c.user_id, c.signup_cohort, c.channel
FROM cohorts c
JOIN deduped e
ON e.user_id = c.user_id
AND e.event_time >= c.signup_time
AND e.event_time < c.signup_time + INTERVAL '7 day'
)
SELECT
c.signup_cohort,
c.channel,
COUNT(DISTINCT c.user_id) AS signups,
COUNT(DISTINCT r.user_id) AS retained_7d,
COUNT(DISTINCT r.user_id) * 1.0 / COUNT(DISTINCT c.user_id) AS retention_7d
FROM cohorts c
LEFT JOIN retained r
ON c.user_id = r.user_id
GROUP BY 1, 2;
Late-arriving events should be handled with a reprocessing window, such as recomputing the last 7–14 days daily. Deduping on a stable key like `user_id + event_name + event_time` prevents inflated retention from duplicate rows.
EasyTechnical
57 practiced
What are the main reasons a team would choose a managed analytics platform like GA4, Adobe Analytics, Mixpanel, or Amplitude instead of building everything from raw events in the warehouse?
Sample Answer
Teams choose managed analytics platforms because they provide speed, consistency, and built-in product analytics features without building everything from scratch in the warehouse.**Main reasons**- **Fast time to value:** dashboards, funnels, cohorts, and retention are available quickly- **Prebuilt identity and session logic:** the platform handles common analytics concepts- **Self-service for analysts/PMs:** non-engineers can explore data without writing SQL- **Lower implementation effort:** SDKs, tag managers, and data layers reduce custom code- **Governance and sharing:** standardized metrics are easier to distribute across teamsA raw warehouse approach is still valuable for flexibility and enterprise modeling, but it usually requires more engineering for ingestion, event modeling, semantic layers, and access controls.As a data engineer, I’d often use both: a managed tool for rapid product insight and the warehouse as the system of record. That gives teams speed at the front end and strong data quality and auditability underneath.
Unlock Full Question Bank
Get access to hundreds of Analytics Platforms and Dashboards interview questions and detailed answers.
