Setting Up Amplitude: A Guide for Product and Engineering Teams

Why Teams Choose Amplitude

Amplitude is particularly strong for teams that need deep behavioural analysis out of the box — its Charts, Pathfinder, and Retention analysis are more accessible to non-technical product managers than raw SQL. If your team needs product analysts to get insights without writing code, Amplitude's UI is an advantage over a pure BigQuery approach.

Installation

For web: npm install @amplitude/analytics-browser. Initialize with your API key and configure the instance:

import * as amplitude from '@amplitude/analytics-browser';
amplitude.init('API_KEY', {
  defaultTracking: { sessions: true, pageViews: true, formInteractions: false }
});

defaultTracking controls Amplitude's auto-capture. For product analytics, disable form interactions auto-tracking (it can capture sensitive field values) and instead track form submissions manually with custom events.

User Identification in Amplitude

Amplitude uses a two-step identification approach similar to Mixpanel:

1. On page load, anonymous users get a random device ID
2. On login/signup, call amplitude.setUserId(user_id) to associate all future events with that user

For cross-device tracking, ensure the same user_id is set on all devices where the user logs in. Amplitude then merges events from all devices under one user profile.

Event and Property Design

Amplitude recommends: event names in past tense ("Report Viewed", "Subscription Started"), user properties in noun form ("Plan Type", "Signup Source"). Set user properties with:

const identifyEvent = new amplitude.Identify();
identifyEvent.set('plan_type', 'professional');
identifyEvent.set('signup_source', 'organic');
amplitude.identify(identifyEvent);

Behavioral Cohorts

One of Amplitude's strongest features is behavioral cohorts — groups of users defined by what they did in the product. Example: users who used Feature X within 7 days of signup. Build cohorts in the Amplitude UI without SQL by selecting events and properties. These cohorts can then be used in retention and conversion analyses to understand how feature usage predicts retention.

Our product analytics service sets up Amplitude with a full event taxonomy and user property schema matched to your product's specific activation and retention metrics. Contact us to discuss your Amplitude implementation.