Traffic Source Analysis

Traffic Source Analysis and Optimization

Understanding which traffic sources generate profit (not just visits) is the key to smart marketing. This comprehensive guide shows you how to analyze traffic sources in Alpha Insights, understand how source detection works, and optimize your marketing budget based on profitability.

Prerequisites: Make sure website tracking is enabled. See Understanding Website Analytics to get started.

Traffic Source Categories

Alpha Insights automatically categorizes all traffic into these sources using the WPD_Traffic_Type class:

Organic Search

What it is: Visitors from search engines (Google, Bing, Yahoo) clicking unpaid search results

How it’s detected: Referrer domain matches known search engine domains:

www.google.com (all country variations)
bing.com
yahoo.com
duckduckgo.com
search.brave.com
ecosia.org
qwant.com
yandex.com / yandex.ru / ya.ru
baidu.com
daum.net
naver.com
Seznam.cz
And 20+ other search engines worldwide

Technical note: Detection checks if referrer contains search engine domain AND typically includes a search query parameter (q=, p=, query=, etc.), though the query itself is not stored.

Typical characteristics:

High intent – actively searching for products/solutions
Good conversion rates (3-8% average)
Higher profit per session
Free traffic (except SEO investment)
Quality varies by keyword and search intent
Often best long-term traffic source

Optimization strategies:

Invest in SEO and content marketing
Optimize product pages for search engines
Target high-intent keywords
Improve site speed and mobile experience
Build quality backlinks
Create helpful, unique content

Google Ads (Paid Search)

What it is: Visitors from Google Ads or Bing Ads paid campaigns

How it’s detected:

First check: Query parameter gclid (Google Click ID) present in URL
Second check: Query parameter utm_medium=cpc from google.com or bing.com
Priority: Query parameters override referrer-based detection

Example URLs:

https://yourstore.com/products?gclid=abc123xyz
https://yourstore.com?utm_source=google&utm_medium=cpc

Typical characteristics:

Very high intent – actively searching, willing to click ads
Excellent conversion rates (5-12% average)
Higher cost per click ($1-10+ depending on industry)
Profitable if managed well (ROAS 3-6 is good)
Scalable – increase budget to increase traffic
Fast results – traffic starts immediately

Optimization:

Tracked automatically via Google Ads integration
Use google_cid={{campaignid}} parameter for campaign attribution
Monitor ROAS (Return on Ad Spend) in Alpha Insights reports
Pause campaigns with ROAS
Scale campaigns with ROAS > 4 (highly profitable)

Social Media (Organic)

What it is: Visitors from social platforms clicking organic (non-paid) posts

Platforms detected:

Facebook: facebook.com, m.facebook.com, l.facebook.com, fb.com
Instagram: instagram.com, l.instagram.com, ig.com
Reddit: reddit.com
TikTok: (detected via utm_source=tiktok)
Pinterest: (detected via utm_source=pinterest)
Twitter/X: (detected via utm_source=twitter or x.com)
LinkedIn: (detected via utm_source=linkedin)

How it’s detected:

Query parameter check first: fbclid (Facebook Click ID) forces Social classification
Query parameter check: utm_source=fb or utm_source=ig or utm_source=Facebook
Referrer domain match: Checks if referrer contains facebook.com, instagram.com, etc.
Social vs Paid Social: If utm_medium=cpc also present, may be classified differently

Technical note: Facebook and Instagram often strip referrer data for privacy. Using fbclid is automatic, but UTM parameters are more reliable for other platforms.

Typical characteristics:

Lower intent – discovery browsing, not active searching
Lower conversion rates (1-3% average)
Younger demographic often (varies by platform)
Free but requires content creation time and effort
Highly variable quality depending on post content
Good for brand awareness and engagement

Optimization strategies:

Create engaging visual content
Use influencer partnerships
Encourage user-generated content
Post consistently with clear CTAs
Test different content types (video, carousel, stories)
Use platform-specific features (Reels, Stories, etc.)

Paid Social (Facebook Ads, Instagram Ads, TikTok Ads)

What it is: Paid advertising on social platforms

How it’s detected:

Facebook/Instagram: fbclid parameter + comes from facebook.com/instagram.com
Manual tagging: utm_medium=cpc or utm_medium=paid_social from social platforms
Campaign tracking: meta_cid={{campaign.id}} for Facebook campaign attribution

Note: Alpha Insights currently classifies Facebook paid traffic as “Social” (not separate “Paid Social”). Use Facebook Ads integration for campaign-level tracking.

Typical characteristics:

Scalable volume – increase budget for more traffic
Moderate conversion rates (2-5% average)
Lower cost per click than paid search ($0.50-$3)
ROAS varies widely (2-5 typical, depends on targeting)
Visual/creative-dependent performance
Good for retargeting and lookalike audiences

Optimization:

Tracked automatically via Facebook Ads integration
Use meta_cid={{campaign.id}} in ad URLs for attribution
Test different audiences and creatives
Monitor profit-based ROAS in Alpha Insights
Retarget cart abandoners and past customers
Use video and carousel ads for products

Email

What it is: Visitors from email campaigns

How it’s detected:

Priority 1 – Query parameters:

mc_cid (Mailchimp campaign ID)
utm_medium=email
utm_source=mailpoet

Priority 2 – Referrer domain matching:

Mailchimp: mailchi.mp, admin.mailchimp.com, campaign-archive.com
Klaviyo: klaviyo.com, klclick.com
Constant Contact: constantcontact.com, r20.rs6.net
Brevo (Sendinblue): r.mailjet.com, mta.brevo.com, sendibm3.com
HubSpot: hubspotlinks.com, hs-analytics.net
Campaign Monitor: createsend.com, cmail*.com (all cmail domains)
ActiveCampaign: emltrk.com, activehosted.com
MailerLite: mlsend.com, emlml.com
SendGrid: sendgrid.net
Salesforce Marketing Cloud: exacttarget.com
SparkPost: sparkpostmail.com
Dotdigital: dotmailer.com

Technical note: Email providers often use tracking domains (e.g., klclick.com for Klaviyo) which Alpha Insights recognizes automatically.

Typical characteristics:

Highest conversion rates (5-15% average)
Very high profit per session
Engaged, existing audience (your list)
Owned channel – don’t pay per click
Excellent for promotions and launches
Repeat customer acquisition

Best practices:

Always use utm_medium=email in all email links
Add utm_campaign to track specific email campaigns
Grow email list aggressively (it’s your most valuable asset)
Segment audiences for personalization
Test subject lines and content
Send abandoned cart recovery emails

Direct

What it is: Visitors typing URL directly, using bookmarks, or unknown source

How it’s detected:

No referrer URL present (empty or null)
Referrer URL is your own domain (internal navigation)
Referrer cannot be determined
Catches all traffic that doesn’t match other categories

Typical characteristics:

Mix of returning customers and brand-aware visitors
High conversion rates (4-10% average)
Often includes some misattributed traffic (missing referrers)
Loyal customers who know your brand
Higher average order value

What “Direct” actually includes:

Typed URL directly into browser
Bookmarks or saved links
Links from native apps (email apps, messaging apps) that strip referrers
HTTPS → HTTP transitions (referrer lost)
Shortened URLs without tracking (bit.ly, etc.)
Links from PDFs or documents
QR codes without tracking parameters
Dark social (private messaging apps)

Note: High direct traffic often indicates strong brand recognition, but some may be misattributed. Use UTM parameters on all external links to reduce “dark traffic.”

Referral

What it is: Visitors clicking links from other websites

How it’s detected:

Referrer URL present and not empty
Referrer domain is NOT your own site
Referrer domain is NOT a known search engine, social platform, or email provider
Catches all external links not classified elsewhere

Examples of referral sources:

Blog features and reviews
News articles and press coverage
Partner websites and affiliate links
Review sites (Trustpilot, Yelp, etc.)
Forums and community sites
Directory listings
Comparison shopping sites

Typical characteristics:

Quality varies dramatically by referring site
Can be very profitable if from relevant, high-quality sites
Conversion rate depends on referring page context and audience
Often untapped opportunity

Optimization:

Identify high-value referral sources in Traffic Channels report
Build relationships with sites that send quality traffic
Create shareable content (infographics, tools, guides)
Pursue guest posting and PR opportunities
Set up affiliate program for relevant partners
Monitor referral traffic for spam or low-quality sources

AI Chat

What it is: Visitors from AI chatbot assistants and search interfaces

How it’s detected: Referrer domain matches AI chat platforms:

Perplexity: perplexity.ai
ChatGPT: chatgpt.com, openai.com
Claude: claude.ai, anthropic.com
Gemini: gemini.google.com, bard.google.com
Grok: grok.com
DeepSeek: deepseek.com

Why it matters: Growing traffic source as users adopt AI search. Different user behavior and intent than traditional search.

Typical characteristics:

Emerging traffic source (small but growing)
High intent if AI recommended your site
Unknown conversion patterns (too new)
May indicate brand authority if AI cites you

Optimization:

Create authoritative, well-structured content
Use schema markup for better AI understanding
Monitor this category – it will grow significantly
Answer common questions comprehensively

App

What it is: Traffic from mobile apps or app-based browsers

How it’s detected: Referrer URL starts with app:// protocol

Use case: If you have a mobile app or app-based shopping, this tracks that traffic separately.

Unknown

What it is: Traffic that couldn’t be classified

When it happens:

Referrer present but doesn’t match any known patterns
Unusual or malformed referrer URLs
New traffic sources not yet in detection list

If you see significant “Unknown” traffic:

Check Website Sessions report to see actual referrer URLs
Add UTM parameters to external links
Contact support to add new traffic sources to detection list

Traffic Source Detection Priority

Alpha Insights checks sources in this order (first match wins):

Query Parameters (highest priority):
- fbclid → Social
- gclid → Google Ads
- mc_cid → Email
- utm_medium=email → Email
- utm_source=mailpoet → Email
- utm_source=fb|ig|Facebook → Social
Referrer Domain Matching:
- Search engines → Organic
- Email providers → Email
- AI chat platforms → AI Chat
- Social platforms → Social
Referrer Analysis:
- Empty/null referrer → Direct
- Own domain → Direct
- app:// protocol → App
- External domain → Referral

Why query parameters take priority: They’re explicit and intentional (you added them), so they’re more reliable than referrer data which can be stripped or missing.

Viewing Traffic Source Data

Traffic Channels Report

The Traffic Channels report is a pre-built report in Alpha Insights. To learn about creating custom traffic reports, see Creating Custom Reports.

Go to Alpha Insights → Website Traffic → Traffic Channels
See comprehensive breakdown of all traffic sources

Data visualizations:

Sessions By Acquisition Channel (Line Chart): Track how each source performs over time
Traffic Data By Acquisition Source (Data Table): Detailed metrics for each source
Traffic By Referral URL: See specific referring domains and pages
Traffic By UTM Campaign: Campaign-level performance
Traffic By Landing Page: Which pages attract which traffic sources

Key table columns explained:

Column	What It Shows	How It’s Calculated
Source	Traffic source category	Determined by WPD_Traffic_Type class
Sessions	Number of visits	Count of unique session_ids
Users	Unique visitors	Count of unique IP addresses
Transactions	Orders from this source	Count of transaction events
Total Value	Revenue generated	Sum of event_value for transactions
Conversion Rate	% of sessions that converted	(Transactions / Sessions) × 100

Source Performance Over Time

Track trends and seasonality:

View the “Sessions By Acquisition Channel” line chart
Identify which sources are growing or declining
Spot seasonal patterns (e.g., email spikes during holidays)
Adjust date range to analyze different time periods

What to look for:

Sudden drops (tracking issues or external changes)
Growth trends (successful optimizations)
Seasonal patterns (plan marketing calendar)
Day-of-week patterns (optimal posting times)

Analyzing Performance by Source

Revenue Per Session – The Key Metric

Formula: Total Revenue / Total Sessions

Why it’s THE most important metric: It tells you the average value of a visitor from each source, accounting for both conversion rate and average order value.

Example analysis:

Source: Organic Search
- Sessions: 5,000
- Revenue: $35,000
- Orders: 225
- Conversion Rate: 4.5%
- Average Order Value: $156
- Revenue per session: $7.00

Source: Facebook Organic
- Sessions: 3,000
- Revenue: $9,000
- Orders: 60
- Conversion Rate: 2.0%
- Average Order Value: $150
- Revenue per session: $3.00

Insight: Organic search visitors generate 2.3x more revenue per session.
Even though AOV is similar, organic has 2.25x better conversion rate.
Recommendation: Invest more in SEO, less in Facebook organic posts.

Profit Per Session – Even Better

In Alpha Insights, you can see profit per session by adding product costs. To add product costs, see Cost of Goods Manager.

Source: Email
- Sessions: 500
- Revenue: $4,500
- Product Costs: $1,800
- Gross Profit: $2,700
- Profit Per Session: $5.40

Source: Social
- Sessions: 3,000
- Revenue: $9,000
- Product Costs: $4,500
- Gross Profit: $4,500
- Profit Per Session: $1.50

Insight: Email generates 3.6x more profit per session.
Even with lower volume, email is far more valuable than social.
Recommendation: Prioritize list growth and email marketing.

Conversion Rate by Source

Shows: Which sources bring higher-quality, ready-to-buy traffic

Example comparison:

Email:           8.2% conversion rate
Organic Search:  4.5%
Google Ads:      6.1%
Social Organic:  1.8%
Direct:          5.5%
Referral:        3.2%

Insights from this data:

Email has highest intent – they’re your customers/subscribers
Google Ads high intent – actively searching for solution
Social low intent – casual browsing, not shopping
Direct high intent – brand loyal, know what they want

Average Order Value by Source

Shows: Which sources bring bigger spenders

Direct:         $95 AOV
Email:          $87 AOV
Organic Search: $78 AOV
Google Ads:     $70 AOV
Paid Social:    $65 AOV
Social Organic: $52 AOV

Why this matters:

Direct visitors often buy more (brand loyal, repeat customers)
Email subscribers tend to spend more (trust relationship)
Social browsers often buy lower-priced items (impulse)
Combine with conversion rate for full picture

Comparing Paid vs Organic Channels

Full Funnel Comparison

Paid Channels (Google Ads, Facebook Ads):

Pros:
- Scalable – increase budget for more traffic
- Fast results – traffic starts immediately
- Precise targeting – reach exact audiences
- Measurable – direct ROI tracking
- Controllable – turn on/off anytime
Cons:
- Ongoing cost – stops when you stop paying
- Profit depends on ROAS – not always profitable
- Increasing costs – CPCs rising over time
- Requires optimization – doesn’t run itself
Best for: Quick growth, launching new products, testing markets

Organic Channels (SEO, Social, Email):

Pros:
- No per-click cost – free traffic
- Compounds over time – grows exponentially
- Higher trust – users prefer organic over ads
- Sustainable – continues even if you pause work
- Higher quality – better conversion rates
Cons:
- Slower results – takes months to build
- Requires consistent effort – content creation, SEO
- Less controllable – can’t “turn up” traffic
- Harder to attribute – indirect conversions
Best for: Long-term sustainable growth, building brand equity

Blended Channel Analysis

Compare total contribution across paid vs organic:

Paid Channels (Google Ads + Facebook Ads):
- Sessions: 4,000
- Revenue: $18,000
- Orders: 180
- Conversion Rate: 4.5%
- Revenue per session: $4.50
- Ad Spend: $6,000
- ROAS: 3.0 ($18,000 / $6,000)
- Net Profit After Ad Spend: $3,000

Organic Channels (Search + Social + Email):
- Sessions: 6,000
- Revenue: $24,000
- Orders: 300
- Conversion Rate: 5.0%
- Revenue per session: $4.00
- Ad Spend: $0
- Net Profit: $10,800 (after COGS)

Insight: Organic channels deliver 3.6x more profit with 50% more sessions.
Paid channels profitable but should not exceed 30% of total traffic.
Recommendation: Maintain paid ads for consistent flow, invest heavily in organic growth.

Source Optimization Strategies

Strategy 1: Double Down on Winners

Identify: Sources with conversion rates and revenue per session above your average

Action plan:

Allocate more marketing budget and time to these sources
Create more content for that channel
Optimize conversion specifically for that audience
Study what makes that traffic convert better

Example: If email has 8% conversion rate and $6.00 revenue per session:

Invest heavily in list growth (popups, lead magnets)
Send more frequent campaigns (test frequency)
Segment list for personalization
Create email-exclusive offers
Build automation flows (abandoned cart, welcome, post-purchase)

Strategy 2: Improve Moderate Performers

Identify: Sources with decent volume but below-average conversion rates or revenue per session

Optimization tactics:

Better targeting: Refine audience to higher-intent visitors
- Organic: Target higher-intent keywords
- Paid: Narrow audience, exclude low converters
- Social: Post content that attracts buyers, not just browsers
Landing page optimization: Create source-specific landing pages
- Match messaging to source (social vs search intent)
- Test different layouts and CTAs
- Reduce friction for that traffic type
Product mix: Promote higher-margin products to this audience
- Feature products that convert better for this source
- Adjust pricing or offers for audience
Retargeting: Set up retargeting for these visitors
- Pixel visitors from low-converting sources
- Retarget with social proof and urgency
- Offer incentive for return visit

Strategy 3: Fix or Cut Losers

Identify: Sources with very low conversion rates or minimal revenue contribution

Decision framework:

If paid channel with ROAS

Try optimization (better targeting, creative, landing pages)

If still unprofitable after 2-3 optimization attempts → Pause

Reallocate budget to profitable channels
If organic channel with

Reduce effort and content creation for that channel

Refocus time on higher-performing channels

Don’t abandon completely – maintain minimal presence
If referral site sending low-quality traffic:
- Check if it’s spam or bot traffic
- Consider requesting link removal if damaging
- Block in robots.txt if malicious

Warning: Don’t cut channels too quickly. Some channels have long conversion windows or assist conversions. Analyze at least 30-90 days of data before deciding.

Device-Based Source Analysis

Combine traffic source with device data for deeper insights:

Facebook + Mobile = 1.5% conversion rate
Facebook + Desktop = 3.2% conversion rate

Insight: Facebook mobile traffic significantly underperforms.
Possible causes:
- Mobile site speed issues
- Mobile UX problems
- Mobile checkout friction
- Audience targeting (mobile users less serious buyers)

Actions:
- Optimize mobile site speed (test with PageSpeed Insights)
- Improve mobile checkout flow
- Add mobile-specific CTAs
- Consider mobile-only landing pages
- Test different ad formats for mobile

Understanding Multi-Touch Customer Journey

Most customers don’t buy on first visit. For detailed information on how sessions and attribution work, see the Session Management guide.

They might:

First visit: Click Facebook Ad (awareness) → Browse, don’t buy
Second visit: Organic Google search (consideration) → Read reviews, compare
Third visit: Direct visit to complete purchase (decision) → Ready to buy

How Alpha Insights handles this:

Session Attribution: Each session is attributed to the source that started it
Order Attribution: Orders are attributed to the landing page and referrer from the order session
Typically last-click attribution: The final session before purchase gets credit

What this means:

Facebook might assist conversions without getting credit
Direct traffic may be overvalued (it’s often the final touchpoint)
Email and organic often get proper credit (high intent, direct conversion)

To understand full journey:

Review individual sessions in Website Sessions report
Track same IP address across multiple sessions
Consider all sources as part of the journey
Don’t undervalue awareness channels (social, content)

Seasonal Source Patterns

Identifying Seasonality

Different sources perform differently by season:

Q4 (Holiday Season - Oct-Dec):
- Direct traffic surges (brand searches, gift shopping)
- Email performs exceptionally (holiday promotions)
- Social organic increases (holiday content sharing)
- Paid ads most expensive but highest volume

Q1 (January-March):
- Organic search increases (New Year resolutions, research)
- Direct traffic drops (post-holiday)
- Email engagement lower (email fatigue)
- Paid ads cheaper (less competition)

Q2 (April-June):
- Social organic peaks (summer content, events)
- Referral traffic increases (spring press coverage)
- Email engagement improves
- Steady growth period

Q3 (July-September):
- Back-to-school drives specific categories
- Traffic generally stable
- Good time for testing and optimization
- Prepare for Q4 campaigns

Use these insights to:

Adjust marketing calendars and content plans
Allocate seasonal budgets (heavy in Q4, lighter in Q1)
Prepare inventory based on seasonal source patterns
Plan campaigns around peak performance times
Test new channels during slow periods

Advanced Analysis: Source × Product Category

Different sources may prefer different products:

Organic Search → Best for:
- Educational products (books, courses)
- Problem-solving products (tools, software)
- High-consideration purchases

Social Media → Best for:
- Visual products (fashion, home decor)
- Impulse buys (under $50)
- Trending items

Email → Best for:
- Repeat purchase products
- New arrivals
- Sale items and promotions

Direct → Best for:
- Replenishment products
- Favorite products
- Subscription items

Action: Analyze traffic source by product category in Report Builder to identify which sources drive sales for which product types. Promote appropriate products to each channel.

Creating Source-Optimized Marketing Plans

You can export traffic data to CSV or Excel for deeper analysis. See Export Features to learn how.

90-Day Marketing Optimization Plan:

Week 1-2: Audit
- Run Traffic Channels report for last 90 days
- Calculate revenue per session for each source
- Identify top 3 performers and bottom 3
Week 3-4: Optimize Winners
- Increase budget/effort on top 3 sources by 50%
- Create more content for these channels
- Optimize conversion paths for these audiences
Week 5-8: Improve Moderates
- A/B test landing pages for medium performers
- Refine targeting and messaging
- Set up retargeting campaigns
Week 9-12: Review and Scale
- Measure improvements
- Scale what’s working
- Pause or reduce what’s not
- Plan next 90 days

Best Practices Summary

Review weekly: Check Traffic Channels report every Monday
Focus on profit per session: Not just traffic volume
Use UTM parameters religiously: Accurate data = better decisions
Don’t ignore low-volume, high-conversion sources: Email with 500 sessions at 8% conversion beats social with 5,000 sessions at 1%
Analyze individual sessions: Use Website Sessions report to understand behavior patterns
Invest in owned channels: Email and SEO compound over time
Use paid channels strategically: Scale proven products, not for testing
Test new sources: Small budgets first, scale winners
Monitor device performance: Mobile vs desktop can be dramatically different
Consider full customer journey: Not just last-click attribution
Track seasonal patterns: Build year-over-year comparisons
Combine with product data: Match products to best-performing sources

Technical Reference: Source Detection Logic

PHP Class: WPD_Traffic_Type

Detection method: determine_traffic_source($referral_url, $query_parameters)

Process flow:

check_query_parameters() – Checks UTM and tracking parameters first
is_traffic_organic() – Matches against $organic_sources array
is_traffic_paid_google() – Currently returns false, relies on gclid parameter
is_traffic_mail() – Matches against $referral_url_email_sources array
is_traffic_ai_chat() – Matches against $referral_url_ai_chat_sources array
is_traffic_social() – Matches against $social_sources array
is_traffic_direct() – Returns true if referrer is empty or null
Final checks: Own domain → Direct, app:// → App, otherwise → Referral

Available traffic types function: WPD_Traffic_Type::available_traffic_types()

Returns array:

[
    'organic'    => 'Organic',
    'google_ads' => 'Google Ads',
    'email'      => 'Email',
    'social'     => 'Social',
    'direct'     => 'Direct',
    'app'        => 'App',
    'referral'   => 'Referral',
    'ai_chat'    => 'AI Chat',
    'unknown'    => 'Unknown'
]

Next Steps

Set up UTM Campaign Tracking for accurate source attribution
Review Website Analytics Overview for tracking fundamentals
Read Session Management to understand attribution windows
Check Troubleshooting Guide if source data seems incorrect
Review Technical Architecture to understand source detection logic
Create custom reports in Report Builder to analyze sources
Set up Facebook Ads Integration for campaign profit tracking
Configure Google Ads Integration for search campaign ROAS

Traffic Source Analysis

Alpha Insights Documentation

Docs Navigation