Science Museum Case Study


Founded in 1857 he Science Museum attracts 3.3 million visitors a year and holds over 300,000 items over five floors. Entry is free but special temporary exhibitions do incur a fee. Visitors include: tourists, families, groups of school children, students, researchers and the general public. In a team of four we were set the task of working in a two week design sprint to design an app that would help visitors find exhibits that would interest them whist helping make their visit much easier and engaging.

The Problem

Due to the large size of the museum visitors can get lost and miss seeing all the items that interest them. As certain exhibitions are very popular crowds and bottle necks can occur wasting visitors time which can their enjoyment of the museum.

The Solution

To provide an app that which allows visitors to successfully plan museum visit seeing as much as possible within their allocated time frame whilst avoiding the crowds.

The Process

Project planning

Our team consisted of Julie, Raphael, Sara and myself. our first job once we had analysed the brief was to set up a project canvas to ensure there was a clear understanding of why the project is being done and what is to be produced. once this was completed we set up the tasks on Trello so maximise our resources and ensure that the objectives were reached.

Project canvas

Competitive Analysis and User Research

We then conducted a competitive analysis of museums and attractions in London, focusing our research on what features they had on their websites and apps and how they made the visitors experience better. Our key findings from this research were that the apps were boring and text heavy, there was little interactive experience and their were no free interactive maps to use for visits. in most cases the map would have needed to be downloaded from the app prior to the visits.

Through or research we managed to identify four key persona groups of tourist, teacher, families and researcher. We then individually conducted user research targeting people we know from these groups. initially we sent out a filter survey via Google forms which helped us find the right candidates for our research.

Our four key persona groups

As a team we conducted a field trip to the Science museum looking to conduct a contextual inquiry on the museum whilst also interviewing people from our four main persona groups.

Science Museum on site research

We then collated the information from our user research and created an affinity map which helped us compile our findings from our multiple interviews and spot trends.

Affinity mapping
Affinity map findings

The key points we pulled out were:

Experience: From learning new information to being able to interact and immerse yourself in an exhibit.

Navigation: A big frustration with our users was being able to navigate around the museum. Many get lost easily, don’t like the big crowds and would like to know how long they should spend in each exhibit.

Purpose: Why the users go to museums: Family outings, a day of escape, their Interests and Visiting a new country.

Planning. There is a mixture of those who like to plan ahead of their visit and book exhibitions early and those who go to the museum on a whim.

Personas and scenarios

Through our user research we were able to create some proto personas, individually we created different scenarios for each of our personas. We decided to focus on our main persona of Margrite who is a married mother of two children who is also a tourist in London wishing to spend a limited amount of time in the science museum.

Margrite — our main persona

Using our main persona of Margrite we began to create scenarios and user experience maps to help us understand potential pain points for our persona.

Creating the experience map for our main persona

Scenario: Margrite has planned to go to the Science Museum with her family, she only has two hours to see as much as possible and visit an exhibition.

Margrites experience map based on our main scenario

Feature Prioritisation

From our findings we used the MOSCOW method to figure out which features to build on. As one of the biggest frustrations our user has is being able to navigate around the museum, this was a big focus for us. So the MUSTS for us were:

To have an interactive map, that uses geolocation to help them navigate. There would also be time suggestions on how long to spend in each exhibit and alerts to tell the user how busy an exhibit is. We also wanted to incorporate a “fun” element in to the app as it is an important aspect to both the brand and user. And finally the ability to have filtering options to choose which exhibits they want to see.

Planning feature prioritization using the MOSCOW method
MOSCOW findings

Design Studio

Once we had decided on the features that we would have on our app we started sketching out ideas of how how we would design those features. This was done in a design studio exercise where we rapidly sketched our ideas based on some scenarios for our main persona Margrite

Our first design studio

In the first session we sketched around the problem the user problem:

1. How can Magritte find her way around the exhibits with her children, and maximise her time in the museum? She has just a few hours.

In the second session we sketched around the problem:

2. How can Magritte find the information she needs to make sure she and her family sees everything the museum has to offer, when her children can’t read English very well? She has a full day.

After that we digested and synthesised the best ideas that would work together as a whole for both problem 1 and then again for problem 2.

We then looked again at both problem and discussed which problem would be of greater value to the user and the client, as well as keeping the overall client brief in mind. The consensus was to move forward with the sketched from problem 1 into a paper prototype.

Initial sketches from design studio

Prototypes and user testing

We then started to design our paper prototypes and test them with users. We initially started with paper, then moved onto a clickable paper prototype. The testing introduced some interesting results which causes us to changes the design of the pages after each round of testing.

Paper prototypes

Paper prototype testing

The sketches were redrawn for mobile and then placed into Marvel.As a team, we tested with users and recorded the findings. The key findings were highlighted and these were used as part of the iteration process.

Clickable paper prototype in Marvel

Some of the findings we found from testing the paper prototypes were: users were going to search bar to search for exhibits - we hadn’t incorporated this functionality in the prototype. It wasn’t clear how the user selected their agenda and planned their day. Users got stuck on certain pages. The burger menu was unneeded as only one option within it and the overall journey not clear.


Clickable wireframe prototype

We incorporated our findings from the paper prototype into a clickable wireframe created in Sketch and in linked into Invision. We then run a second set of user testing on the wireframes, our finding were; Add walking distances to the map, add heat map to the floorplan, make agenda editable, make the book exhibition more prominent change some of the call to actions to be more friendly and remove search bar.

Mid fidelity prototype

We then took the findings from our user testing on the wireframe prototypes and added them to our mid fidelity prototype, we then conducted further user testing to finalise our design and move into a high fidelity prototype.

Mid fidelity prototype
Iterations of the designs through testing

Conclusions and next steps

After finalising our two week design sprint we looked at some of the features we could add to the next steps in the apps development.

As the world is gradually shifting into a new way of experiencing digital we believe that at the science museum our smart phones can become the tools to make this happen.

Scan Options:
Scanning through the activation of a camera view point at an object and receive a form of interaction with the subject ahead.

Play feature — Scavenger hunt:
The play feature entices visitors to connect with learning at the museum through the app. A proposed game scenario would include an active hunt that draws all users and owners of a smartphone to participate in. The museum could make some by allowing them to build their knowledge ratings

Language Options:
As the Science museums footfall is heavily driven by tourist, exploring in your native tongue may add more value to your overall experience and understanding of the content available on show at the museum. More on permitting visitors to extract a higher level of knowledge from their trip to the Science.

It would also bridge the gap between existing gadgets and media at the museum that foreigners can not make sense of because the language is only presented in English.

Social Media:
Allow for users to share their personal experience at the science museum. To friends and family so that they too can have a similar experience. Allowing for groups to discuss what they saw.

3D Visualisation:
Allow for the standards of the App to be taken up a notch in sophistication. And become a more visually pleasing to the user.

One clap, two clap, three clap, forty?

By clapping more or less, you can signal to us which stories really stand out.