NRMA Insurance
Digital car insurance - conversion optimisation
Customer Problem
For customers wanting to do multiple car insurance quotes especially when shopping for a new car, it was observed that it was a pain point for people to have to re-enter all of their personal details and required information multiple times to compare quotes. Customers were also often starting their quotes in the digital channel and found themselves needing to call the call centres for either reassurance in their answers or help with understanding certain questions.
Our overall assumption was that if we; introduced more conversational language and made it easy for customers to get multiple car quotes by 'remembering' their details they would have an improved and more efficient quoting experience resulting in increased customer engagement and increased conversion rates.
Business Objectives
-
An increase in conversion rate and GWP
-
An increase in our CAS score for consider, purchase and reconsider
-
An increase in usage of NRMA Online Account
-
an increase in Digital Channel Share
Team
-
Lead UX Designer
-
Mid UX Designer / researcher - Me
-
Channel Manager, Product Owner, IM, UI Designer, 2 x developers, as well as
-
A broader Stakeholder group including Digital Channel and Sales EM's (involved in weekly playback sessions.)
Design Approach
We followed the double diamond approach over an initial four week period, allowing each one week for Discover, Define, and Deliver. This was followed by multiple weeks of experimentation with coded prototypes, to test and learn.
Discover
-
We collected all existing research available within the company including:
-
Sales journey qualitative research
-
Sales pathway analytics and qualitative data
-
CASS scores and customer research
-
Competitive analysis
-
Customer journey maps and personas
-
Research wall
customer sales journey, sales quote data, IAG personas, competitive analysis
After conducting an analysis of all of the information gathered, we did some collaborative grouping, theming and synthesising of our findings to create 24 key insights.
6 of the key insights
Comparisons
"I research online so that I can use comparisons to give me confidence in my decision"
Make it easy
"30% of customers get quotes for multiple vehicles and find it frustrating having to fill out the form again for each one"
Know me
"Know me enough to provide a tailored offer, and use your expertise"
I want control
"I research online so that I have control. To properly understand the information & make a decision in my own time"
Value me
"I want to be valued as an existing customer because I've earned it"
Access anywhere
"I use multiple devices in all different contexts to do my research"
​
Define
We were fortunate enough to have had a dedicated Tesla project room for the duration of this work. During the define phase we explored all of the key insights that were previously discovered. This was done in a very hands on environment and approach.
​
We ran multiple workshops in our project room throughout this phase starting with creating opportunities and taking stakeholders through the process of how to craft "How might We" (HMW) statements, collaboratively working through some rapid idea generation rounds using post it notes which were then ran through a synthesis and prioritisation process.
​
Both myself as the Mid-UX Designer as well as the Lead Designer on this project took turns in the workshop facilitation and worked together in developing the understanding of the Design Thinking processes that we were involving and leading our stakeholders through.
The prioritisation for this piece of work on the most desirable / important HMWs to ideate on were created via involving various types of business owners and decision makers including deign to do some zen voting. The opportunities with the highest votes were the ones taken into rapid sketching for ideation.
​
By having all key decision makers involved we were able to get the perspectives, buy-in and priorities from all sides of the business and customer problems.
​
​
Opportunities workshop
Identifying problems & opportunities from insights
With the 24 key insights outlined and printed upon the wall Some key decision makers from the wider stakeholder group joined us to collaboratively identify the problems Which were turned into opportunities as HMW statements.
6 Ideas prioritised for design
​
1. Display an estimated price range at the top of the quote which gets more narrow / accurate as the quote becomes more and more completed.
2. A sticky banner at the top of the quote to display % of quote completion. This progress bar would be aimed at setting expectations to elevate frustration in not knowing how long the process will take.
3. A conversational messaging bot - via Facebook Messenger to ask questions and give customers a quote., in a similar way the call centre staff would interact with our customers.
​
4. Enable customers to change their vehicle on the quote / price page. Making it possible for people to get multiple quotes for different vehicles without having to redo the quote flow.
​
5. Display the share button on the quote summary page (for joint decision makers)
​
6.Reduce the amount of effort and time needed by customers by reducing the steps, using smart defaults, and reducing the question set.
Design
To kick off the design phase, myself and the lead UX designer on this project both worked through the ideas and proposed solutions that had come out of the rapid sketch sessions by creating an experiment board to catalogue the most highly voted on / prioritised ideas on a spreadsheet.
Here we were able to align it back to the problem it was solving and create the success criteria that each idea would have to met during the validation process.
​
Concepts for experimentation
click to zoom
While in this phase I predominately further fleshed out the concepts while the lead UX designer worked on creating interactive Axure prototypes to take to testing.
​
The first round of validation testing was actually done as gorilla testing with NRMA customers at the Parramatta branch with a quick prototype that a developer worked with us to develop.
​
The two primary objectives in our first round of testing was to understand how our customers felt about the addition of "happy copy" to the quote flow, a more conversational style of feedback to answers given within our usually formal and jargon-filled insurance app, which was one approach we were considering for the Facebook Bot quote concept.
As well as a feature on the final price page, which allowed people to "change vehicle" with one entry field to see all of their quote / estimates update to reflect the different car type/model they were wanting a quote for.
working prototype
click to zoom
The happy copy received mixed feelings and feedback as customers needed to feel a sense of authority, expertise and trust in the process of buying insurance.
While the change vehicle feature was highly desired it required further exploration as more needs were indicated through the testing such as, people would then expect to be able tp save each quote. individually to come back and refer to them later or share with joint decision makers.
​
We continued to do another 5 rounds of customer validation testing over the next 5 weeks.
The cycle was set up to use the week prior, to simultaneously create each concept in Axure while the testing was being planned.
​
For this. process, I worked closely with the lead designer who was executing the prototypes, while I was recruiting the test participants, creating the test plans, with our objectives, assumptions, and success metrics, as well as organising the test scripts and note takers process; and consequently after each round there was a I led the synthesis of findings, created test reports and the business play backs to keep all stakeholders across our findings and progress.
​
Our findings from each round allowed us to further enhance and iterate on our design while implementing a new feature or idea to be validated for the subsequent round.
​
These rounds of validation testing took place in our internal test labs, which had 2 rooms, one for facilitation with a participant and the other room next door displayed and recorded each session.
5 concepts - for testing
click on images to zoom
Rd1 Concept
single question, shopping cart, onboarding
Rd 4 Concept
saved quotes / shopping cart
Rd 2 Concept
multi-quote flow,
Face Book quote bot
Rd 5 Concept
Easier quote entry
Rd 3 Concept
easy buy flow
Some of these concepts which were created and iterated on, don't align perfectly to our initial experiment board, as our findings through the validation period meant that our ideas and insights were further explored and enhanced as the process evolved.
​
Overall we were very fortunate to be able to be so authentic with our design thinking process and continued to stay very true to our customers true needs and wants while maintaining continued buy-in from our main business owner - the digital sales channel manager.
Deliver
While the value of this project was well established and highly desired from stakeholders such as digital sales channel managers, and digital EGMs.
Unfortunately, the business did not endorse this piece of work for funding due to higher-level priorities in technical re-platforming investments.
​
However, this piece of work including the research and findings were diligently documented and often referenced in projects that the product and design teams undertake to this day.