Experiment: Increase Conversion of High Intent Visitors before Site abandonment and reducing drop off

​

Overview

Currently we receive more than half a million website visitors and our current conversion rates stands at 5.5%+ - 6% which way better considering competition benchmarks in our industry.

But there is another factor that about 90 - 95% of our website visitors don't convert in their first visit they need time as it's a consultative focused industry so users usually take time to make a purchase, typically 2 - 4 days of return visit as reordered in our Data platforms.

How can we do that?

There are two major ways to do it at scale
- Reducing drop offs and Bounce rates
- Retarget users with ads & CRM

Challenges

Retargeting often gets results it's costly and already in practice

Increasing the First part data collection can help but in our previous set of experiments we have noticed that use of aggressive popups often increases the CPM's across platforms, increases user drop offs more if it miss fires ( mostly 75% of the time happens ) + as most of them were casual browsing users converting them via popups were often difficult and would result in high burn thorough CRM channels as they had low intent.

To make popups engaging we tried spin the wheel and other sort of stuffs but the intent and conversion were not good in that case as well normal popup has been working well against them.

So idea was very clear to use popups but with intent based approach in showcasing them rather than randomly showing the popup to everyone.

Plus we as a brand don't support aggressive discounting so we have to factor in that as well

Observation

We found through our systems that Homepage is often the first source of landing page for returning users and validated it with few sample customer calls, so we ran few popups

Hypothesis

If we showcase Popups to high intent based funnel actions like Add to cart, Initiate checkout but not purchased then we could increase our Conversion rates of abandon visitors to purchase would increase from 28% current to 35%

Goal

The Goal of the experiment is to reduce the drop off in conversion by Site abandoning users and converting them which would lead to increase in revenue impact directly and first part data capturing with for high intent users rather than collecting data of casually browsing users


Success Metrics

Increase the Visit to conversion rate from 28% to 37% for Returning abandon visitors

Best Possible - 37%
Least Possible - 32%

Experimentation Design

What are we testing -

We are testing if a popup with a discount code could increase the site visitors who have initiated a checkout but due to some reason they are leaving the site to complete the order in their first session

What are we not testing

implementation of popup to every casual browsing user on the site with a discount

Variation Design

This test would have one variation and one control ( Normal homepage with no popups )

Variation - A

​

Audience ( Size & Segmentations )

Audience - Anonymous users who have initiated the checkout but haven't completed the order and are about to leave the website.

Sample Size required for experimentation

Sample Size per variation - Optimizely

Sample Size per variation - VWO

​

Before Launch - Testing

​

Post Experimentation

Experiment Results - Variation A out performed control in terms of conversion and reduction in drop off

Control - Homepage with non discount popup for anonymous visitors with action ( Initiate checkout but order not placed )

Variation A - Popup with 25% OFF

​

​

Release Decision - Scaling it as it's performing better than control

Learning

Discounts when given to anonymous users with a higher intent to buy have higher chances of conversion and stickiness rather than discounts given to random low intent users

Next Steps​

While the above test was perfect and came in our favor to increase the conversion further and target high inter users we could create more filtered segment testing like
- Users who add to cart but no checkout or purchase
- Users who browsed more than 2 pages before going through checkout process
- Users who have already purchased from us and dropping off after some action
- Testing the popup with adding more info elements like Email + Phone number

Stakeholder Management

​

​

​