Project Overview
Lead the migration, as well as an enhancement on the existing preschool search feature from ECDA’s existing website, to LifeSG’s web and app platforms.
My Contributions
User Research, User Testing interviews, User Interface Design, Frontend Handoff
About the Team
1 Intern designer, 8 - 10 Engineers (Multiple stacks)
.png)
To build a one-stop gateway for all of National Library Board (NLB)'s vast wealth of resources, to position NLB as Singapore's Knowledge Resource.
As the lead designer behind the new pre-school search and enrolment features on the LifeSG app, the objective was to simplify and enhance the user experience for parents looking for pre-schools for their children.
This project was both a migration, as well as an enhancement on the existing preschool search feature from ECDA’s existing website, to LifeSG’s web and app platforms.
The preschool search feature services more than 200,000 parents and couples every year in Singapore looking for a preschool for their children, Boasting comprehensive listings of over 1,900 licensed and registered pre-schools, including those run by the Ministry of Education. This ensures parents have seamless access to a wide variety of options.
Designed and prototyped the experience for ECDA revamp and migration to LifeSG, both on web and mobile devices.
Prepared, recruited and conducted 15 user testing sessions based on personas required by stakeholders. User testing sessions involved usability testing based on prototype, A/B testing, SUS scoring and profile specific questions.
Communicated findings to stakeholders and relevant product owners within GovTech.
Handoff design to different tech stacks for development — front-end and back-end developers were required for different integration of features.
Issue one — Inconsistent preschool search experience
Before the enhancement and migration of ECDA preschool search to LifeSG, their desktop experience functioned within their web domain, while their mobile experience was only available on LifeSG. Thus burdening the user with poor cross-platform experience.
After the enhancement, parents can now freely search for pre-schools on both the LifeSG website, mobile web and LifeSG app, creating a seamless and consistent experience. This makes it easier to find conveniently located pre-schools. The search can be further refined based on criteria such as pre-school level, availability, fees, mother tongue language, food options, and transportation services.
Issue two — Lack of personalised search
Initially, past searches were not able to be saved, thus requiring users to repeat their search inputs repeatedly for every session.
As an enhancement, we leveraged on information within a user’s Singpass to generate personalised search results — information such as residential location, number of children, child’s age and education levels were used to provide more accurate search results for users.
Thus, eliminating the need for users to repeatedly key in the same information in each returning session.
Issue three — Inaccurate information
ECDA serves as a ‘broker of information’ for preschools. Thus, it has limited authority in enforcing preschools to update crucial details such as vacancy availabilities of classes.
This caused much frustration as there would be a mismatch in expectation from parents on the available slots and reality of the slots a preschool has, due to outdated data.
However, after the revamp, preschools are required to provide more up-to-date information on details such as vacancy availabilities, fees, and contact details if they were to continue using ECDA as a platform.
NLB had prior User Experience case studies done by consultancies, which our team used to value add and validate our user research analysing NLB’s web experience.
Through our research, we identified two main pain-points of the website:
Based on our primary research and pain points from existing users, the focus was to design a seamless experience across multiple platforms for the feature. After which, using the prototype, we validated the design though a series of 15, in-person user testing sessions and brought back the comments for further synthesis and refinement.
The process was research intensive as we had specific pain points we had to solve — however, we also referenced commercial booking websites such as AirBnb, Booking.com and skyscanner to ensure that users would not require much effort to navigate within our search feature.
2 months • (September - October 2023)
Research and discovery
Studying the existing ECDA website before migration, and understanding the profile of users through informal knowledge gathering by speaking to parents within our social to learn about their needs.
1 month • (October - November 2023)
Design and ideation
Our team prototyped and redesigned the features based on our research, leveraging on existing design systems used across several clusters within GovTech, mainly LifeSG as well as BookingSG to expedite the design phase.
2 months • (November - January 2023)
User testing and validation
Based on several profiles for our feature, we began to craft our user testing discussion guide and proposals. This is a comprehensive document containing the objectives of the user tests, questions and breakdown of the profiles we would be interviewing. Simultaneously, we had to recruit and schedule user testing sessions.
1 month • (January - February 2024)
Synthesis and refinement
Using Dovetail, we extracted key findings based on the 15 in-person interviews -using them to refine our design to be used as an ideal state, as well as begin the discussion for MVP state. These findings were then presented to our ECDA stakeholders for feedback and approval.
The objective of this feature is to enable couple or parents to search and plan for which preschool they would want to send their child, seamlessly.
The main features we developed was in response to the current pain points experienced by users. Using these insights, we began to enhance the features.
Issue 1: Addressing the lack of an inconsistent search experience
As LifeSG front-end engine is based on React, we are required to translate the mobile experience onto web. By utilising the same code base on both web and mobile platforms, it opened up the possibility for us to introduce more features to enhance the seamlessness between devices.
Quality of life improvements
On top of having seamless a multi-device experience, visual enhancements were also made to the unoptimised web experience ECDA used to have. Using a consistent design language and interaction enabled areas such as responsiveness and accessibility to improve.
We introduced a bookmarking feature which was accessible on any device. This proved to be a welcomed enhancement as initially, user’s bookmarks only existed either on ECDA’s website, or the LifeSG app — but these were not in sync.
We were also able to enhance the sorting and search logic of the search — initially, the search on ECDA’s website was a combination of a list and map search. As a result, sorting and filtering of preschools were limited as they had to adopt the logic of both a map and list view.
Specific audiences for different result views
We then separated the search into a list and a map view, enabling different filters to be applied to the search. During our user testing sessions, we made the observation that these two views also served two different groups of profiles.
Users who used the list view were generally individuals who were at the start of their preschool search journey. As a result, they were more inclined to view multiple preschools at once, casting their nets wider to explore their options.
Filters such as sorting were of greater use for individuals looking to view a greater number of preschools.
On the other hand, once users have narrowed down their selections, they generally used the map view as it helps them visualise the preschools around their preferred location. Users who are searching for preschools with a location they are unfamiliar with (e.g. in-laws place) would also tend to use the map view to get a grasp of the preschool location.
Thus filters such as the vacancy availability of preschools around their location would prove to be of greater use for individuals.
Issue 2: Lack of personalised search experience
With the feature now hosted on LifeSG, we were able to leverage on the use of Singpass to retrieve their Myinfo data (with their consent).
Information such as their address and child’s age were values that could be detected and passed into the filters to simplify a parent’s search for preschools. As these details are tied to their profile, they need not re-enter their details for subsequent uses of the feature.
Based on the child's age as well as starting NRIC characters, we are able to infer the child's citizenship as well as preschool level, reducing the number of details required to be filled during search.
Issue 3: Inaccurate information
One predominant issue parents faced while searching for preschools was the lack of updated information. Information pertaining to preschool vacancies are crucial for parents as demand for certain services such as infant care are often competitive.
As mentioned earlier, ECDA acts as a ‘middleman’ for preschool data — ECDA has limited power to enforce preschools to update their vacancy information. However, with the migration for preschools to host their information on ECDA’s platform, preschools have to be more proactive in updating details such as fees, vacancy availabilities and contact details.
An enhancement to provide parents with more visibility of preschool availabilities was to enable them to view preschool availabilities 6 months into the future — enabling parents greater ability to plan and shortlist preschools for their child.
Parents are able to search for the specific month they intend on sending their child to preschool. As a result, the search results will show which are preschools with availabilities during that month.
Validation of our design and assumptions are an integral part of the design process in GovTech. User testing and synthesis of our findings took up majority of the effort, but was equally fulfilling. We interviewed a total of 15 participants from the public, across different socio-economic and family backgrounds.
Dovetail was used to catalogue and organise our recordings and transcripts — after which, we categorised our findings into several general buckets based on our design objectives.
Discussion guides
Creation of discussion guide to document the objectives of user testing, our target interview profiles, the validations to the assumptions made during the design, the types of tests we would be conducting (A/B tests, sorting, click tests…) and the types of questions we would be asking.
Recruitment survey
Next would be the recruitment survey, which we would create to send en-mass to members of the public. Here, we would filter out their responses and choose a range of users to participate in our user testing. Some of our filters include: number of children, income per household, marital status, whether they are living with their parents, etc.
Synthesis
After the 15 participants have been interviewed, we begin to synthesise the data based on several metrics. This list is non-exhaustive, and it includes buckets such as:
a. Behavioural and personal preferences
b. Page feedbacks and interaction
c. Nice to have features or enhancements
d. Existing pain points
While this project was a step outside my comfort zone, with none of my team members being parents — I throughly enjoyed the various aspects of user testing and validation as it enabled me to speak with a plethora of people from all walks of life.
I wrote a detailed post capturing the insights of my journey.
Confession: User testing is terrifying
This project was extremely fulfilling as I had a part to play in building Singapore's learning resource and to inspire a new generation of users who would look to the National Library as Singapore's knowledge resource.
During this project, I also had the opportunity to learn how a web interface can have such a strong impact in the perception and the positioning of an organisation. Little details and decisions made with stakeholders ended up shaping this project to be much greater than originally envisioned.
