Product History

ClearScore Protect was a product that lived within the ClearScore App but was very separate from the normal credit space proposition as it focused more on identity fraud and had a subscription model to it. It is a product that was very early in its lifecycle and was still considered a MVP. It had some great impact and reached 2 Million Activations and 20K paid subscribers which was a strong showing of what it was offering users

Post Discovery

After a 5 week discovery period which ended in ideation workshops and a new sets of opportunities areas we had a large amount of work in front of us. Lots of this research and many discussions we had looked at how we could create more value for our users and move Protect into a better version of itself but there were still a few aspects we needed to nail before developing solutions

At this stage we had uncovered a range of opportunity areas, it was now up to me and the other designer in my squad further discover more insights and to start developing concepts

The Evaluation Pillar

Internally we referred to opportunity areas as pillars so people understood what we were speaking of in terms of solution development. The first area we wanted to look into was evaluation, we felt like this would address a range of different user problems and be beneficial for the product’s activation goals. We kept getting a range of ideas from people within the company since Protect launched to do more in the product that had this type of feedback mechanism and more importantly our discovery pointed at some core user problems to solve in this area too

We looked at a range of different solution types and decided on going with a “Security Rating” solution as a direction. We had quite a bit of buy-in from initial chats with Product / Engineering around solution types like this and we felt like it would have the brand impact Protect needed

Responsibilities

I was responsible for problem space exploration and solution development from beginning to end. I led on discovery and problem space exploration and designed the end solution in collaboration with the other designer in the team in order to meet tight timeframes. We also kept C-level and department heads in the loop on this work on a weekly basis

Next Steps - More Discovery

There was an exception at this stage for design in our squad to be jumping straight into developing solutions but after a review of this work the other designer and I agreed that one of us should dive into the problem area a bit more while the other should focus on some of the short term delivery for now. This resulted in me looking at the ‘evaluation’ opportunity area more

Discovery Concluded

At this stage we had gained enough about our users needs, attitudes and behaviours and had a better understand of industry learnings. We also knew more about the risks that users could encounter, what data that could power their experiences, what it means for them and what actions can be done from our product currently and going forward

We defined the following key problems and opportunities:

• User’s didn’t correlate online behaviour / footprint to fraud

• User’s had an elevated sense of safety

• We could analyse data sources and highlight fraud risks to people in a more human manner

• We better understood what actions equal lower fraud risk

“How Might We present an evaluation which highlights fraud risks to our users in order to encourage them to develop safer security habits?”

In this step, we wanted to generate ideas off of how we could represent this security information to the user. From the research we did on feedback mechanisms, we understood that there were different ways to display evaluation to our users and we set off to explore them

These were the main ideas that came out of some ideation from myself and the other designer in the squad. We had reviewed many of the solution types and tried to output them in different ways. The purpose of this was to explore many different paths to a solution and see what fitted the best. The scored evaluation felt more on brand compared with the other types since ClearScore is known for disclosing a Credit Score. It is likely that users would have more of a mental model for driving action

We decided on a score approach because it allowed us to display information to our users in a digestible and interactive way. Our learnings from ClearScore and OneScore also gave us strong signal that a score helps users understand their overall situation and motivates them to take action. This seemed to fit our vision of giving simple evaluation and developing better security habits

In order to integrate this new evaluation type, we looked at how we could evolve the base product and how we could shift the IA forward based off of the audit of the current product. We did this to show how we could wrap the product around this evaluation feature

I explored further solution types that were focused on providing information and incentivising user action through gamification. This concept wasn’t taken further at this stage as the team decided to focus on the score evaluation first

Designing a score the best way

At this stage we knew a range of things about the problem area and had decided on a solution type but now it was about designing it the right way. We put together a list of learnings we wanted specifically about scores before we started development of the solution

A Smaller “Score” Discovery

The whole core team was involved from this point onwards very heavily. We started a ‘Score Framing’ discovery to gather important learnings and insights on how to develop a number based score the right way

Expert Interviews - OneScore

In the company there was a team who in the past designed an app that had a score (and quite a complex one). We interviewed the Lead UX designer and Head of Product that were involved with this project and spoke with them. We learnt a lot from their journey building a score, and even though their business model and product was far from Protect, it helped highlight areas of opportunity we hadn’t thought about. They’ve also helped us defined constraints and issues we were likely to run into

Competitor Analysis

I personally had done this before in the older discovery so didn’t spend too long here but we aimed at looking at how different companies displayed scores

Questions that should be solved by the solution

At this point we had enough of what we needed to now start developing the solution with real confidence. We highlighted the following questions to be answered:

1. How does the score benefit the user

2. How does the score provide ongoing value

3. How does the score reflect the users safety

4. How does the score get informed by different data sources

5. How should be build the right user experiences off of it

6. How do we promote friendly comparison (Should we even do this)

7. How does the score change from free and paid (Or should it be the same)

8. How do we make sure that this score isn’t confused with the Credit Score

Splitting development up

We were unlucky in that we didn’t get to speak with any design or product experts that had launched successful freemium model businesses but we were successful on finding areas that we wanted to further research if we got the opportunity . Doing this exercise as a design pair helped us further research the right things and refine our thinking on tests that we would want to run in the future

Driving an Experience with a Score

I did quite a lot of thinking about what aspects of the user experience could power the score. While the score was going to be mainly informed by data sources, I wanted to understand how we can then turn static data sources into something easy to communicate that would be relevant and tailored to users needs. This piece of thinking I’m really proud of but we opted to “park” it for now until V2, where we could fully integrate this thinking

Calculation Type - Bounded Score

This piece of work was led by the analyst and over a series of group calls we agreed on how the score would be calculated from a technical point of view. This stream of work was happening alongside the UX development of the score from a design POV so there were many calls over 2 weeks to nail this in a manner that would work across the board. We decided on a bounded score type and it would be out of 100. This way it would better match the mental model of the credit score but be a 2 digit number to show a difference

Derisking the Score Calculation

Whilst going through the score calculation we asked many questions from a UX point of view of how certain edge cases might play out. We had many things to take forward in the design which helped us focus on what to deliver for a V1

Early Distribution

The analyst was able to map different forms of the model to our current user base. We used this to nail the weightings of the different categories (action types)

Usability Testing - A Basic Score

I drove a lot of testing with the score early on (while we were still figuring out how it would be calculated). We had many questions this solution needed to answer and I focused on trying to understand how users would perceive the score

I began with a quick usability test with a basic score design on top of the current Protect experience, the test had several questions some which were multiple choice and some open questions in which I noted down mostly what they said and what they did when interacting with the design

Interestingly people who saw it had a range of different thoughts on what the score was, how it was calculated, what order they pressed core categories, what it was measuring against and it fundamentally changed how they perceived actions

Usability Testing - A Basic Score + Drilled in view

So after the first test I learned a lot of things and fed this back to the team. I spent a few days going through the results and iterating on the solution to address some of the issues highlighted in our test results

Main aspects I iterated was the language and user interface to focus the user more around the product offering. I added a newer page which I called the “drilled in” page which would show to the user how their score had changed and how they could improve it. For the main overview I had tried to add more context to the design and added more branding elements to it. I then set up a test with 5 people to understand similar goals to the first test and note what changed

More folks this time understood the proposition and showed more interest in “pressing” elements. They overall had more questions and had a more accurate understanding of how the product worked

Usability Testing - A Basic Score + Score Analysis + Explainer

I had spent some time iterating the older design off of the feedback I received to This one was more visually applying and I tried to derisk score reception in a more direct way by adding another page (the dark one on the right)

I didn’t test this design with users but did guerrilla testing with staff and had feedback from my squad and design team

Focus Shift

So at this stage the other designer I was working with started focusing on the visual design of the score and how it would sit in the overview page. On my side, I focused my efforts on how we would communicate the score to users, how it worked and why actioning security actions could benefit them

Core Designs

These show the core designs we delivered to change the app going forward. This shows “core” pages across large and small sizes as the designs themselves are heavily dynamic. I will cover off how these journeys work and different states later

Section Detail Pages AKA “Sandwich pages”

These pages were visually similar but changed depending on user progress, subscription tier and category type