adeyemi@adediranadeyemi.com +234 816 273 5399
User Behavior · Path Analysis

Mapping Where Users Convert — and Where They Don't

A deep-dive path analysis of 9,935 sessions on a subscription learning platform — identifying the exact conversion sequences that lead to paid subscriptions and the drop-off points that silently kill revenue.

Tools
Python · Pandas · Google Analytics · SQL
Industry
EdTech · SaaS · Subscription
Type
Journey Analysis · Funnel Analytics · Feature Importance
Customer journey analysis showing user conversion paths and drop-off points on a subscription platform
9,935 User sessions analyzed
3,798 Login interactions — most frequent touchpoint
1,773 Direct-to-checkout conversions
27.9% Feature importance of login page for conversion

Project Overview

Analyzing user behavior is essential for understanding how users engage with a platform and identifying opportunities for improvement. This project examines user journeys on 365's subscription-based learning platform across 9,935 sessions — mapping the paths from first touchpoint to conversion (or abandonment), and identifying what differentiates users who subscribe from those who don't.

The analysis spans user journey sequences, start and end point distributions, conversion path rankings, subscription-type behavioral differences, machine-learning-derived feature importance, and a structured comparison of converting vs. non-converting user behavior.

Central finding: The most frequent interaction on the platform is logging in — and sessions that start and end at login, without progressing to pricing or checkout, represent the platform's single largest missed conversion opportunity. Coupons and checkout page interactions are the strongest conversion drivers.

Objectives

The primary objectives of this analysis are:

  • Map the most common user journey sequences across the platform and identify which paths lead to conversion
  • Identify start and end points that indicate high-intent vs. disengaged behavior
  • Rank conversion paths by frequency to prioritize UX optimization efforts
  • Segment behavior by subscription type (Annual, Monthly, Quarterly)
  • Quantify the feature importance of each page interaction in predicting conversion
  • Identify drop-off points where users disengage without converting
  • Deliver actionable recommendations to increase conversion rates and reduce churn

Data Dictionary

The dataset contains 9,935 rows and 4 columns, capturing session-level user journey sequences:

ColumnDescription
user_idUnique identifier for each user on the platform
session_idUnique identifier for each user session
subscription_typeThe type of subscription the user holds: Annual, Monthly, or Quarterly
user_journeyA sequence of pages visited by the user during their session — the primary analytical variable

Dataset note: The dataset captures page-visit sequences but not session duration, time-on-page, or demographic attributes. These limitations are addressed in the Limitations section and inform the Further Research recommendations.

Analysis

The analysis was structured across seven interconnected layers — from broad interaction counts down to the behavioral differences between users who converted and those who didn't. Each layer builds on the previous to progressively narrow the focus toward actionable insight.

User Journey Insights

Top Interactions Across the Platform

Across all 9,935 sessions, five interactions dominate total engagement volume:

Log in
3,798
Homepage
2,396 sessions
2,396
Checkout
2,021 sessions
2,021
Sign up
1,210
Coupon
1,041 sessions
1,041

Login dominance alert: Login appearing as both the most common start point and end point of sessions indicates a large segment of users who return to the platform repeatedly but never progress beyond basic re-authentication. These are users with genuine interest but no clear path forward — the highest-leverage retention opportunity.

Start Points & End Points

Session Start Points

Where users enter the platform most often:

Start PointSessions
Log in2,379
Homepage2,329
Checkout1,776

Session End Points

Where sessions most often terminate:

End PointSessions
Log in3,601
Checkout1,964
Coupon1,029
Sign up883

The gap between login as a start point (2,379) and login as an end point (3,601) reveals that many sessions begin elsewhere but terminate at login — users hit a friction point and fall back to the login page rather than progressing forward. This is a strong signal of navigation confusion or a lack of compelling next-step prompts after authentication.

Conversion Paths

The top conversion paths reveal a striking pattern: the overwhelming majority of conversions happen through a single direct step — landing straight on checkout. Multi-step paths involving Pricing are a distant second.

Checkout
1,773
Pricing Checkout
72
Homepage Pricing Checkout
68
Courses Pricing Checkout
7
Other Pricing Checkout
3

Key implication: 96% of conversions that go through a multi-step path flow through the Pricing page. This makes the Pricing page the single most critical conversion lever on the platform — any friction, confusion, or weak value proposition there directly suppresses conversion rates.

Coupon Influence on Conversions

Coupon interactions appear in 1,041 sessions and end-points in 1,029 sessions — indicating that coupons are not just used mid-journey but are often the final action before a user exits or converts. Users who encounter a coupon are significantly more likely to proceed to checkout, making promotional campaigns a direct conversion driver.

Subscription Type Analysis

Behavior varies meaningfully across the three subscription tiers — each showing a distinct engagement pattern:

Annual Subscribers

Engage most frequently with Login and Coupon steps — indicating loyal, returning users who are motivated by discounts. Longer journeys with deeper platform engagement.

Monthly Subscribers

Exhibit shorter, direct engagement patterns — navigating directly to checkout with fewer intermediate steps. Higher purchase intent on entry, lower exploration behavior.

Quarterly Subscribers

Similar to monthly in directness, but slightly more likely to visit the pricing page before committing. Mid-tier consideration behavior — comparing value before converting.

Strategic implication: Annual subscribers are the loyalty cohort — they respond to discounts and re-engage frequently. Monthly and quarterly subscribers are intent-driven — they arrive ready to buy and need friction removed, not persuasion added.

Feature Importance for Conversion

A machine learning model was trained to predict conversion from page interaction signals. The feature importance scores reveal which page visits carry the most predictive weight:

visited_Log in
0.2793
visited_Other
0.2013
visited_Coupon
0.1910

Interpretation

  • Login (27.9% importance): The login page visit is the strongest single predictor of conversion — but in a nuanced way. It indicates a returning user with prior intent. The challenge is that many login sessions end without progressing forward.
  • Other pages (20.1% importance): The "Other" category aggregates less-common page visits — their collective importance suggests that exploratory behavior (browsing beyond the core funnel) is a meaningful conversion signal.
  • Coupon (19.1% importance): Coupon interaction is the third strongest conversion predictor — confirming that promotional offers are a critical conversion lever, not a nice-to-have feature.

Actionable insight: The top three conversion predictors — login, coupon, and exploratory browsing — all suggest that the highest-converting users are returning, discount-responsive, and curious. Retention campaigns, personalized coupon offers, and content discovery features would directly target this behavioral profile.

Drop-Off Points

Understanding where users exit without converting is as important as understanding where they do convert. The analysis identified several high-volume drop-off patterns:

Login as Exit Point

3,601 sessions terminate at the login page — the highest drop-off volume of any page. Users re-authenticate but receive no compelling next step to guide them forward.

Blog Page Drop-Off

Sessions terminating at the Blog reflect disengagement after consuming informational content. Users read content but find no path to action — no CTA, no personalized recommendation, no next step.

Resources Center Drop-Off

Similar to Blog — informational pages attract users but fail to channel them toward higher-value actions. These pages convert passive browsers into passive exits.

Homepage Bounce

A large number of non-converting users terminate at the Homepage without progressing to Pricing or Checkout — suggesting weak value proposition communication or unclear navigation hierarchy.

Converted vs. Non-Converted Users

The behavioral contrast between users who subscribed and those who didn't is stark — and directly actionable:

✓ Converted Users

Behavioral Profile

  • Shorter journeys with focused, high-value page interactions
  • Disproportionately high checkout and pricing page visits
  • Frequent coupon interactions — discount-responsive
  • Return users (login as start point) with clear intent
  • Less time on exploratory or informational content
✗ Non-Converted Users

Behavioral Profile

  • Longer or more fragmented journeys without reaching checkout
  • Frequent sessions starting and ending at login without progression
  • Engagement concentrated on Blog, Resources Center, and Homepage
  • Less coupon interaction — missing the discount trigger
  • High bounce rate from informational pages without conversion CTA

The fundamental difference is intent clarity. Converted users arrive knowing what they want — a subscription — and navigate directly toward it. Non-converted users explore the platform's content but never receive a sufficiently compelling reason, at the right moment, to commit.

Recommendations

1

Optimize the Post-Login Experience

Since login is both the most common start and end point, the post-login landing experience is the single highest-leverage intervention point. Implement personalized dashboards, progress reminders, or course recommendations that guide users toward pricing or checkout within seconds of logging in. Remove dead-ends.

2

Enhance the Pricing Page

96% of multi-step conversions flow through the Pricing page. Audit the page for clarity, load speed, and value proposition strength. Test social proof elements (number of subscribers, course completion stats), comparison tables, and prominent CTAs. Even small improvements here have outsized conversion impact.

3

Leverage Coupons as a Conversion Trigger

Coupon interactions are the third strongest conversion predictor (19.1% feature importance). Expand targeted promotional campaigns — offer personalized discounts triggered by specific behavioral signals: 3+ login sessions without checkout, blog consumption without progression, or return visits to the pricing page.

4

Add CTAs to Blog and Resources Pages

Sessions terminating at Blog and Resources Center represent users with genuine interest who lack a bridge to conversion. Embed contextually relevant CTAs — "Start learning this topic now" → Pricing — or inline course recommendations within blog posts. Convert passive reading into active exploration.

5

Tailor Strategies by Subscription Type

Annual subscribers respond to loyalty incentives and discounts — target them with renewal reminders and exclusive coupon offers. Monthly and quarterly subscribers have high purchase intent on entry — remove friction from their direct path to checkout rather than adding persuasive content that slows them down.

6

Intervene Earlier in Non-Converting Journeys

Non-converting users tend to disengage after browsing low-value pages. Implement exit-intent triggers on Blog and Resources pages, personalized email follow-ups after multiple sessions without conversion, and behavioral retargeting campaigns targeting users who visited Pricing but did not reach Checkout.

Python Pandas Path Analysis Google Analytics SQL Conversion Optimization User Journey Funnel Analytics SaaS Analytics Feature Importance

Limitations & Further Research

Current Limitations

  • No session duration data: The dataset captures page sequences but not time spent — making it impossible to distinguish between deep engagement and rapid skimming within the same page sequence.
  • No demographic information: Age, location, gender, and device type are absent — all of which are known moderators of digital subscription behavior.
  • No new vs. returning user differentiation: New visitors and returning users are analyzed together, despite having fundamentally different intent signals and optimal conversion strategies.

Recommended Further Research

  • Session duration analysis: Add time-on-page and session length data to understand the correlation between engagement depth and conversion likelihood.
  • Demographic segmentation: Incorporate user demographics to identify whether specific age groups, geographies, or device types have meaningfully different conversion patterns.
  • Behavioral cohort analysis: Segment users by behavioral patterns — frequent returners vs. one-time visitors — to design precision retention strategies for each group.
  • A/B testing framework: Instrument the recommendations above (post-login CTAs, Pricing page improvements, Blog CTAs) with A/B tests to measure lift and prioritize future development investment.

Work with Adediran Adeyemi

Do you know where your users drop off — and why?

I map customer journeys, identify conversion killers, and turn session data into decisions your product team can act on. First call is free.