Growth Experiment - A/B testing
Growth Experiment - A/B testing
Growth Experiment - A/B testing
Bowtie's official website homepage growth experiment by conducting a/b testing, successfully increased conversion rate by 238%
Bowtie's official website homepage growth experiment by conducting a/b testing, successfully increased conversion rate by 238%
Bowtie's official website homepage growth experiment by conducting a/b testing, successfully increased conversion rate by 238%
Company
Bowtie Life Insurance
Company
Bowtie Life Insurance
Company
Bowtie Life Insurance
Tools
Google Analytics
Google Tag Manager
Google Optimize
Webflow
Hotjar
Tools
Google Analytics
Google Tag Manager
Google Optimize
Webflow
Hotjar
Tools
Google Analytics
Google Tag Manager
Google Optimize
Webflow
Hotjar
Role
Project owner
Product designer
Web developer
Role
Project owner
Product designer
Web developer
Role
Project owner
Product designer
Web developer
Deliverables
Website revamp design
A/B testing result & analytics
Deliverables
Website revamp design
A/B testing result & analytics
Deliverables
Website revamp design
A/B testing result & analytics
Overview
As a product designer, I worked closely with growth team and other design team members to improve conversion rate optimization (CRO) in order to boost business performance through website, given that website is the first touch point with our customers. A good homepage can play a crucial role in brand building, feeding useful product information, educating customers how bowtie solves their problem in life and most importantly encourage user to learn more and buy our products.
That’s the reason why I purposed conducting A/B testing on the most visited page - homepage, to presumably make the largest impact on business performance. In the experiment period, we followed 4 main steps including - observation, hypothesis, run experiment and evaluate the result.
Overview
As a product designer, I worked closely with growth team and other design team members to improve conversion rate optimization (CRO) in order to boost business performance through website, given that website is the first touch point with our customers. A good homepage can play a crucial role in brand building, feeding useful product information, educating customers how bowtie solves their problem in life and most importantly encourage user to learn more and buy our products.
That’s the reason why I purposed conducting A/B testing on the most visited page - homepage, to presumably make the largest impact on business performance. In the experiment period, we followed 4 main steps including - observation, hypothesis, run experiment and evaluate the result.
Overview
As a product designer, I worked closely with growth team and other design team members to improve conversion rate optimization (CRO) in order to boost business performance through website, given that website is the first touch point with our customers. A good homepage can play a crucial role in brand building, feeding useful product information, educating customers how bowtie solves their problem in life and most importantly encourage user to learn more and buy our products.
That’s the reason why I purposed conducting A/B testing on the most visited page - homepage, to presumably make the largest impact on business performance. In the experiment period, we followed 4 main steps including - observation, hypothesis, run experiment and evaluate the result.
Step 1 - Observation
Customer enquiries analysis - Top 3 insights
Our UX designer created 2019 Q2 analysis on customer enquiries to get insights about what users concern the most
People Do Not Know Us Well.
(CHI: 你地屬於咩公司？你資金點來？你地做咗幾內？你地結業我點算？)People Want To Know How Premium Increases.
(CHI: 索償會唔會增加我嘅保費？)People Have Concerns with How our Application & Claim Works.
(CHI: 入院時有無幫手？點找數？投保程序係點？)
Google Analytics result - Due to confidential reason, all data & numbers will not be revealed in the portfolio
Second highest traffic landing page
Other than quantitative research on users feedback, we checked Google Analytics (GA) and observed that second highest traffic landing page is the home page, while the top one is VHIS Flexi plan product page.
Step 1 - Observation
Customer enquiries analysis - Top 3 insights
Our UX designer created 2019 Q2 analysis on customer enquiries to get insights about what users concern the most
People Do Not Know Us Well.
(CHI: 你地屬於咩公司？你資金點來？你地做咗幾內？你地結業我點算？)People Want To Know How Premium Increases.
(CHI: 索償會唔會增加我嘅保費？)People Have Concerns with How our Application & Claim Works.
(CHI: 入院時有無幫手？點找數？投保程序係點？)
Second highest traffic landing page
Other than quantitative research on users feedback, we checked Google Analytics (GA) and observed that second highest traffic landing page is the home page, while the top one is VHIS Flexi plan product page.
Step 1 - Observation
Customer enquiries analysis - Top 3 insights
Our UX designer created 2019 Q2 analysis on customer enquiries to get insights about what users concern the most
People Do Not Know Us Well.
(CHI: 你地屬於咩公司？你資金點來？你地做咗幾內？你地結業我點算？)People Want To Know How Premium Increases.
(CHI: 索償會唔會增加我嘅保費？)People Have Concerns with How our Application & Claim Works.
(CHI: 入院時有無幫手？點找數？投保程序係點？)
Second highest traffic landing page
Other than quantitative research on users feedback, we checked Google Analytics (GA) and observed that second highest traffic landing page is the home page, while the top one is VHIS Flexi plan product page.
Step 2 - Hypothesis
If the flow and content of homepage is changed, the performance and conversion rate will increase because the copy and flow are more feature-focused and customer-centric.
Step 2 - Hypothesis
If the flow and content of homepage is changed, the performance and conversion rate will increase because the copy and flow are more feature-focused and customer-centric.
Step 2 - Hypothesis
If the flow and content of homepage is changed, the performance and conversion rate will increase because the copy and flow are more feature-focused and customer-centric.
Step 3 - Run experiment
Subject: Homepage
Period: 20 days (Sep 12 - Oct 1, 2019)
Total Traffic: 4,197 Sessions
Homepage 1.0 (Old)
Traffic: 2,015 (~50%)
Homepage 2.0 (New)
Traffic: 2,182 (~50%)
What is the difference between Homepage 1.0 and 2.0?
Build trust
Since people do not know us well from the research which means that they do not trust a new start-up when buying insurance product, I re-designed the version 2 by adding many build-trust component, such as testimonials, cyber-security proof, media coverage and the photo of our senior advisor John Tsang, who is a well-known person around Hong Kong society.
Amplify service experience
People have concern about how online insurance work. They do not know if there is any service for an online platform. That is why service experience section has been placed to ensure customers that bowtie will fully take care of them not only online, but also offline.
More sale-oriented
I put two options in the top section to take care 2 different groups of customers -
1. who wants to buy fully by themselves online, and
2. who wants customer service to guide them through buying process.
We also make sure that the entire process is not time consuming -
1. 10 seconds to get quotation online, and
2. customer service will get back to customers within 48 hours.
Step 3 - Run experiment
Subject: Homepage
Period: 20 days (Sep 12 - Oct 1, 2019)
Total Traffic: 4,197 Sessions
Traffic: 2,015 (~50%)
Traffic: 2,182 (~50%)
What is the difference between Homepage 1.0 and 2.0?
Build trust
Since people do not know us well from the research which means that they do not trust a new start-up when buying insurance product, I re-designed the version 2 by adding many build-trust component, such as testimonials, cyber-security proof, media coverage and the photo of our senior advisor John Tsang, who is a well-known person around Hong Kong society.
Amplify service experience
People have concern about how online insurance work. They do not know if there is any service for an online platform. That is why service experience section has been placed to ensure customers that bowtie will fully take care of them not only online, but also offline.
More sale-oriented
I put two options in the top section to take care 2 different groups of customers -
1. who wants to buy fully by themselves online, and
2. who wants customer service to guide them through buying process.
We also make sure that the entire process is not time consuming -
1. 10 seconds to get quotation online, and
2. customer service will get back to customers within 48 hours.
Step 3 - Run experiment
Subject: Homepage
Period: 20 days (Sep 12 - Oct 1, 2019)
Total Traffic: 4,197 Sessions
Traffic: 2,015 (~50%)
Traffic: 2,182 (~50%)
What is the difference between Homepage 1.0 and 2.0?
Build trust
Since people do not know us well from the research which means that they do not trust a new start-up when buying insurance product, I re-designed the version 2 by adding many build-trust component, such as testimonials, cyber-security proof, media coverage and the photo of our senior advisor John Tsang, who is a well-known person around Hong Kong society.
Amplify service experience
People have concern about how online insurance work. They do not know if there is any service for an online platform. That is why service experience section has been placed to ensure customers that bowtie will fully take care of them not only online, but also offline.
More sale-oriented
I put two options in the top section to take care 2 different groups of customers -
1. who wants to buy fully by themselves online, and
2. who wants customer service to guide them through buying process.
We also make sure that the entire process is not time consuming -
1. 10 seconds to get quotation online, and
2. customer service will get back to customers within 48 hours.
Step 4 - Result
We got really positive result in homepage 2.0 at the end - longer reading time, creating more engagement, less bounce rate. And most importantly, from a growth perspective, homepage 2.0 created more lead, more policy application started and more policy application submitted.
A/B testing result summary
Metrics we used to measure the result
Google analytics trend in 20 days
Step 4 - Result
We got really positive result in homepage 2.0 at the end - longer reading time, creating more engagement, less bounce rate. And most importantly, from a growth perspective, homepage 2.0 created more lead, more policy application started and more policy application submitted.
Step 4 - Result
We got really positive result in homepage 2.0 at the end - longer reading time, creating more engagement, less bounce rate. And most importantly, from a growth perspective, homepage 2.0 created more lead, more policy application started and more policy application submitted.