Untangling information around pensions.

Talia is an early star-up aiming to close the gender pension gap, which is currently double the gender pay gap.

Pensions are usually passive, boring and a difficult topic to talk about for most people.
Talia’s mission is to change all this and make pensions accessible, easy to use and understand, to empower people, and especially women, to be confident when saving for their retirement.

My role as Lead Designer was to deliver the first MVP for the seed round. The role also included logo design, branding, and consulting on positioning and go to market strategy


Today, women retire 40% poorer than men. This is because pensions don’t take in consideration different lifestyle, care responsibilities and career breaks.


| HMW change this percentage? HMW make pension less boring and more engaging?


To empower women to take ownership of their long-term financial future without feeling as though they’re impacting their quality of life today

Our Process


During discovery we learned about motherhood penalty, the appalling misinformation around pensions and the gender pension gap. 
Most people assume that the reason behind the gender pension gap is because of the disparity of wages between men and women – with the pension contribution being a percentage of your salary, if you earn less, you can save less – but this is not the only reason. Most women will take, at some point in their career, some sort of break from the workforce, for having children or caring for elderly in their family.
Also, after having children, many women will go back to work part-time, usually without a pension. It is been calculated that, on average, a woman loses 5 years of contribution per child.

As a first step to dig deeper in this topic we talked with women in maternity leave, pregnant or with small children. We prepared a loose questionnaire to be sure to cover all the topics we wanted. The main focus was to understand how much these women were aware of the motherhood penalty, how they  were feeling about motherhood and how their employer was treating them during this journey.

Some of the questions we asked

Were you aware of the gender pension gap?

What arrangements did you make before going into maternity leave?

Is pension something you think about? How often?

Do you know how much is in your pension pot?

What are your plans for returning to work?

We recorded all session and used Dovetail to organise our research. On the transcript we tagged crucial parts to find patterns and common pain points.

“I am too busy figuring out maternity allowance! I don’t have the brain capacity to think about my pension”

“I am constantly looking after another human being. Pension is the last of my problem right now.”

“I have a rough idea of how much I have saved, but I am not too bothered at the moment.”


What we learned from these interviews was that pregnant women and women in maternity leave were not our target to start with. They were not at all interested in pension and too busy with all things related to the upcoming tiny human or too busy trying to keep that tiny human alive and maintain some sanity.
Most of the women were concerned about being paid the right amount either via maternity pay or allowance.

We realised we could make a bigger impact targeting 20-30 years old women without children. Our assumption was that having enough knowledge around pension and learning about the pension gender gap, would make them think more about their pension and take action earlier – and this in return will actually make the biggest impact because of compound interest.

New target personas


Crazy 8s

Big Ideas

For the product ideation part of our journey we run a workshop. We included pension and financial experts and product designers to ensure to have the right mix of creativity and feasibility. We wanted as much different backgrounds as we could, to allow different perspectives to arise and get all the ideas that we could. It was a very intense experience, but it was amazing to see so many competent and intelligent people coming together trying to solve such an important problem.

Some slides from our workshop

testing the winning ideas

From the workshop we had 7 concepts we wanted to test. In order to test them as quickly as possible I mocked up some low fidelity prototypes using just Procreate and we showed them to 13 women. We scored the results against 6 metrics: Trust, Ease of user, Usefulness, Amount of money, Frequency of deposit (days), Money now/later.

The winning concept was the cashback concept. Easily add up to your pension pot while shopping.

After testing the app concept we continued using Maze to test the UI using Click heat maps and unmoderated tests with different tasks. We also tested high level prototypes to ensure the interactions were clear enough. We tried to incorporate some gamification on the Grow screen, to prompt people to save more money trying to complete their goal and unlock the bonus.
I really liked this phase because we were able to solve problems using our creativity and the immediate feedback from unmoderated testing allowed us to continuously test and iterate.


We then got accepted into the BGV incubator, which meant we needed to sat down to decide what we wanted to achieve at the end of it and what we wanted to present to investors.

There were 2 main roads: keep investigating the financial product and ask for a bigger investment to make it happen, or starting with a smaller MVP and smaller investment and then build up investment from user traction. We picked the second route.
The first step of this new phase was to define what we could build with the least effort that could bring the biggest value to our user.

USER flows

Sign up / Log in

Design & Test

In terms of design we took a iterative approach, with continuous testing of the different screens and flow. We used a combination of unmoderated and moderated testing using different platforms and techniques like Maze, Userinterviews, UsabilityHub (now Lyssna) and Zoom.


During the discovery phase, both for product and branding we used Maze and UsabilityHub to quickly test desirability, usability, colours, layout, copy, emotional responses, and many others variables with different tools like survey, prototype testing, 5 seconds tests, first click tests and preference tests. This allowed us to check and iterate quickly with a sizeable amount of users different concepts and designs.
One thing we immediately discover from testing was that pink was working well for branding, but not so much for the app interface. Almost all users didn’t trust the product, but when we then switched to green the trust metrics grew from 5% to 65% – that was the moment we decided to split our colour palette between branding and product.


For the onboarding we tested multiple parts, other the usual drop off points, unclear steps and design errors we tested for three specific insights. As a first test we wanted to test a longer onboarding. We assumed that our user would have not liked to land on an empty screen, so we tried to collect as much information as possible during the onboarding, but as expected all users told us the onboarding was too long and they’d have to see some value before committing to this kind of process.

We then tested email onboarding versus telephone number and the majority of users agreed that they felt more comfortable using their email address since it feels less personal.

Lastly we tested our conversational UI, to try and make the process feel more human and less transactional, but our users were not familiar with it, so they mentioned that they could’t understand it and also they’d prefer a shorter onboarding experience. 


The first design for the Lessons took inspiration from Instagram stories. We wanted our user completely focused on the 5 minutes lesson, so we made sure to eliminate the rest of the interface and make the lesson full screen, with a different background and a very standard layout, with scrollable content and input field when needed. Users could also exit the lesson and save progress if needed.
When tested, we found out that users were not engaged, they would scroll, without thinking the content was making any difference. This is when we realised we were falling down the same path of other pension providers: we were not talking to our users, we were showing them information. This was not in line with Talia brand, it needed to feel more human and more like a conversation. So we designed a new UI, a conversational one – it was not a real conversation, because we didn’t have any chatbot, but it felt way more like a 2 way conversation. When we tested the new design we realised we made the right choice: all the users liked the conversational UI and appreciated that it felt more liek talking to a friend giving you advices instead of a Company telling you how things works.


We immediately realised the amazing opportunity to make the variables informative and align them with the lessons: we linked a lesson to each of the variables where we could collect the data while informing the users on how the variables would affect the final numbers.

We tested the first UI iteration of this screen and we encountered multiples problems: the users told us they did not understand how the part below would affect the numbers at the top. They felt it was not visually engaging and a bit boring. It also felt a bit crammed.

After this discovery we explored multiple other option and we tested what we though were the best options. The final UI scored a lot better than the first one: The users immediately connected the variables to the numbers at the top and had a visual indication (both checkmark and inputted data) of the completeness of each variables. They could see immediately which one where still need to be completed and which one were completed already.

Information Architecture


We worked with an external agency to build the Talia app.

We decided to go for React for speed and maintainability. We then tested on both iOs and Android alongside development.


I developed the Design system while building the MVP – each new component went up to build on the brand new system, so when we had to create new components I had a clear idea of what we already had into the product and could be re-used and what needed to be created from scratch. This also helped a lot to maintain a consistent style and look throughout the screens. Further development would include a purge of unnecessary components, a better organisation of components and an icon rules definition.

This was obviuosly a first stab to the Design System, if I were to develop Talia more, I would audit the screens to check for component and interaction consistency and try to keep the creation of new elements on the low side.


Talea (tà-le-a) is the Italian word for scion, which perfectly represent pension contribution. From a small severed part, with time, patience, and for pension, some compound interest, a whole new plant can grow.
We tested the name alongside with some others to test for friendliness, freshness and mental association.
We used Maze to conduct these tests.
Talia resonated well with users, especially in terms of approachability, friendliness and trust.
When we tested the Talia logo we even got 3 people saying it made them think of pensions. Jackpot! 🎉


Focus on the problem

As a product and UX designer I always struggled with the suggestion to “soak in the problem”. I’ve always been a solution-driven designer, but this time, because we had to come up with the solution from nothing, or, to be more specific, just from the problem and user’s frustration, we really soaked in the problem and it made it easier when we had to reduce our scope for our MVP.
The lack of knowledge was a massive opportunity to bring actual value to our users and help them to better understand pensions and investments.

Creating a Business model is tough

While working on this projects we had multiple discussion
on which Business model was best suited for Talia, which one would have been the most sustainable and which one felt the correct one to us.
We though about B2B2C, but couldn’t find a way to make it premium without feeling like we were betraying the values who started the whole project.
We then took the decision to go directly B2C, but I think, in your early stages as a start-up founder it will always feels unsure. I think it is the essence of early days, some gambles pay off and some don’t.


When we started Talia we had good data backing up our assumptions of where we could focus to fix the gender pension gap, but it was clear as light that we were wrong after we interviewed the women we wanted to focus on. This was an invaluable lesson about qualitative data and their power when designing with empathy. 

This reminded us that assumptions are just assumptions, and you should always look for data to back them up while being open to change your assumptions.

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from - Youtube
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google