Introduction
ThinkLouder is a B2C company specializing in Scrum and Agile training courses. The client sought to gain insights into various aspects of their business but faced challenges due to a lack of tracking mechanisms on their custom-built website.
Challenges
During the discovery call, the client was unable to answer several key business questions due to the absence of actionable data tracking. The questions included:
- Which landing page variations drive the most conversions and for which marketing platforms?
- For which instructor are most registrations occurring?
- Which training course is the most popular among users?
- How many times do users visit the website before signing up for a course?
- How many days from the first visit does it take for users to sign up?
- How many days before the course date do users purchase the training?
- For which course types (weekend or weekday) and what times of the month (beginning, middle, end) do users prefer?
Strategy
To address these questions, a comprehensive data tracking strategy was developed, involving:
- Landing Page Tracking: Code to store landing page variations in a 1st party cookie via JavaScript.
- DataLayer Implementation:
- Course Name as Item Name
- Training Dates as Item Variant
- Training Timings as Item Category 2
- Instructor Name as Item Category
- Traffic Source Tracking: Stored in a 1st party cookie via JavaScript.
- Session Count Tracking: Calculated and stored in a 1st party cookie via JavaScript.
These strategies were implemented through Google Tag Manager, with DataLayer integrated into the website. Various funnel events/steps were tracked, including View Item, Add to Cart, View Cart, Begin Checkout, Add Payment Info, and Purchase.
Transformation
After implementing the tracking strategy and performing rigorous testing, the data accuracy was 93% when reconciled with backend CRM data. Data collected over 45 days was exported to an SQL-based database, where the following steps were taken:
- Un-nested parameters such as item category, item name, and item variant.
- Standardized parameters like item variant (training dates) and purchase date for mathematical analysis.
- Stored 1st party cookie data for session duration and integrated it with funnel events/steps.
- Created sub-tables to address each business question.
- Established routines to run SQL queries periodically for data processing.
Outcome
The analysis provided the following insights:
- Landing Page Variations: Pages 8 and 10c were most effective, with 8 performing best for Facebook ads and 10c for Google ads.
- Signup Timing: 70% of users signed up between 0 to 7 days before the training.
- Training Date Preference: 72% of users preferred training dates in the middle of the month.
- Instructor Performance: 4 out of 21 trainers generated 90% of the revenue.
- Training Preferences: 87% of users preferred weekend afternoon/evening sessions.
- User Behavior: On average, users visited the website 4.7 times over 32 days before making a purchase.
Impact
The insights led to actionable changes:
- Increased training sessions in the middle of the month.
- Focused marketing efforts on weekend training.
- Initiated a remarketing campaign for upcoming courses within the next 7 days.
These adjustments contributed to a 9.7% average monthly growth in the client’s business.