Web & Marketing Analytics¶

Web and marketing analytics encompasses the tools and methods for measuring website effectiveness, traffic source performance, and cross-channel attribution. The discipline combines analytics platforms (GA, Yandex Metrica), tag management (GTM), UTM tracking, and attribution modeling.

Key Facts¶

Analytics is not a numbers profession - it's a way of thinking about causal chains
Web analyst tasks: set up data collection, build reports on KPIs, analyze traffic sources, analyze funnel, form and test hypotheses
UTM tags are the standard for traffic source tracking in URLs
Google Tag Manager eliminates developer dependency for tracking setup
Cross-channel (end-to-end) analytics connects ad spend with actual revenue across all channels
Attribution models determine how credit is distributed across touchpoints

Patterns¶

Key Web Metrics¶

Metric	Description
ROI/ROMI	Return on advertising investment
Revenue	Total income
Average Order Value	Mean transaction size
LTV	Revenue over user's lifetime
Retention Rate	% returning users
Bounce Rate	% sessions with single pageview / no interaction
Pages per Session	Content consumption depth
Conversion Rate (CR)	% completing target action

UTM Tags¶

URL parameters for traffic source tracking:

site.com/?utm_source=google&utm_medium=cpc&utm_campaign=summer_sale

Parameter	Purpose	Examples
`utm_source`	Traffic source	google, yandex, vk
`utm_medium`	Channel type	cpc, email, banner
`utm_campaign`	Campaign name	summer_sale
`utm_term`	Keyword (search)	running_shoes
`utm_content`	Ad variant (A/B)	blue_button

Web Analytics Tools¶

Google Analytics (GA4): event-based tracking, audience/acquisition/behavior/conversion reports, goals and funnels, segments and custom dimensions. BigQuery export for raw data.

Yandex Metrica: session replay (Webvisor), click/scroll/movement heatmaps, form analytics, funnels, segments. Deep integration with Yandex Direct.

Hotjar: heatmaps, session recordings, feedback polls, conversion funnels. UX research focused.

Google Tag Manager (GTM)¶

Tag management system - no developer needed for tracking code: - Triggers: page view, DOM ready, click, scroll, form submit, custom event - Variables: CSS selectors, data layer values, URL parts - Preview and debug mode for testing before publish

DataLayer push example:

dataLayer.push({
  'event': 'purchase',
  'ecommerce': {
    'transaction_id': 'T12345',
    'value': 25.42,
    'currency': 'USD'
  }
});

Traffic Source Categories¶

Organic search (SEO)
Paid search (PPC/contextual advertising)
Social media (organic + paid)
Email marketing
Direct traffic
Referral traffic
Display advertising

Attribution Models¶

Model	Credit Distribution	Use Case
Last click	100% to last touchpoint	Default, most common
First click	100% to first touchpoint	Discovery channel analysis
Linear	Equal across all touchpoints	Fair overview
Time decay	More credit to recent	Consideration-heavy products
Position-based	40/20/40 first/middle/last	Balanced first+last emphasis

Cross-Channel Analytics¶

End-to-end analytics = connecting all data sources: - CRM (deals, revenue) - Ad platforms (Google Ads, Yandex Direct, Facebook) - Web analytics (GA, Metrica) - Call tracking (phone leads) - Email platforms

Gotchas¶

GA4 is event-based (not session-based like Universal Analytics) - this fundamentally changes how data is structured
UTM parameters are case-sensitive - utm_source=Google and utm_source=google create separate entries
Bounce rate definition differs between GA versions and should not be compared across tools
Attribution models can give dramatically different results for the same data - always understand which model is in use
Webvisor/session recordings raise privacy concerns - ensure GDPR/privacy compliance in your region