Data Analysis

Analyzing Ingenium Museums
Website

Ingenium
Ingenium is a corporation overseeing three Canadian national museums: the Canada Agriculture and Food Museum (CAFM), the Canada Aviation and Space Museum (CASM), and the Canada Science and Technology Museum (CSTM). Ingenium came to our team of four to examine the websites' performance indicators identified by Ingenium as being of interest to them: referral traffic, user path/content flow, and time on page, in order to gain more insight into how the sites were performing.
MY ROLE

My roles included:
- Analyzing the KPIs of the client sites' Google Analytics
- Creating mock-up recommendations and writing report
- Presenting final presentation deck

TEAM

Our team of four worked collaboratively to submit a written report and presentation as our final deliverable to Ingenium.

PROJECT DURATION

Summer 2021, 4 Weeks

What was the problem?

Our client

Ingenium sites homepage

Ingenium in Canada

Ingenium is both a corporation and a brand. As a corporation, it oversees three Canadian national museums: the Canada Agriculture and Food Museum (CAFM), the Canada Aviation and Space Museum (CASM), and the Canada Science and Technology Museum (CSTM).

The Problems

Analyzing enormous data in the previous 5 years

After an initial meeting and follow-up communication, our team knew that our client particularly would like to gain insight into its website performance for each of their websites (Ingenium, CASM, CAFM, and CSTM) for the period 26 June, 2017 to present.

The data comprised of two different versions of design

The corporation began using Google Analytics in 2017 and the data derived from that platform forms the basis for this current analysis. The corporation completed a redesign of its web properties in 2020 in which various websites were split into individual entities. The redesign introduces a complicating factor when examining the data across the five-year span of the study.

Method of Anaylsis

Process

1

2

3

Defining the goals

After meeting with the client, the research team focused on carrying out a study of website performance, especially in term of content, for each of the websites (Ingenium, CASM, CAFM, and CSTM) for the period 26 June, 2017 to present.

Chose the appropriate metrics according to the goals

The team examined this time horizon using performance indicators identified by Ingenium as being of interest to them: referral traffic, user path/content flow, and time on page. The team also investigated new users, bounce rate, and exit rate in order to gain more insight into how the sites were performing. 

Created Dashboards in Google Data Studio

The team created four dashboards, where were designed for this analysis and set up for client to monitor and analysis the site performance metrics.
Ingenium’s web properties underwent a redesign that was completed in 2020. As a result, the team chose to employ a before-and-after comparison in the design of the dashboard.

The Four Dashboards

We created the following four dashboards specific for the goals of this study: a Before VS After Redesign Dashboard, a Monthly Dashboard, a User Journey Dashboard, and a User Behavior Dashboard.

Monthly Dashboard

Before VS After Dashboard

User Journey Dashboard

User Behavior Dashboard

Limitations of the Study

1. Misrepresentation of Audience

2. Only Quantitative Data

Google Analytics is not fine-grained for categories such as gender but, instead, solely relies on a male/female dyad for all gender assignments. This may mis- or over-represent binary gender categories while mis- or under-representing non-binary gender categories.

Google Analytics provides copious amounts of quantitative data that reflects what users do but does not provide qualitative data regarding why they do it.

3. Unclear Impact of the Design Changes

4. Unclear Impacts from Covid-19

Without explicit information regarding the exact changes made in the redesign process, and an A/B comparison analysis, there is insufficient data to gauge what impact, if any, a particular change may or may not have had.

The roll out of the redesign during the global COVID-19 Pandemic also has injected a large degree of uncertainty into the performance numbers as it is unclear how, exactly, the closure of the various museum properties as well as changes to browsing and information-seeking behavior may have impacted the performance of each site.

Analysis Results

Overall Findings

There are shared patterns that are worth noting but each site has individual characteristics and patterns that must also be taken into account. The recommendations provided strive to recognize the tension of similarity and difference inherent in this type of composite property. In addition, it is also apparent that the closures caused by the COVID-19 pandemic had an effect on the museum’s traffic.

4 Major Findings

Based on the analysis of Ingenium’s web properties, the Pratt team focused on 4 key dimensions: shared patterns, individual and shared areas of strength, areas in need of redesign, and social media patterns. Each of these is discussed in greater detail below. In addition, we end this report with avenues for future consideration.

1

Shared Patterns

2

Individual & shared strengths

3

Areas Needed Redesign

4

Social Media Patterns

1. Shared Patterns

The data show cyclical spikes in traffic in March and July and dips in December and January. These likely correspond to the rhythms of the school year. 

These spikes appear to be quite predictable and it is advisable to take them into account for yearly planning purposes.

Recommendations

1. Take advantage of the March and July spikes, it is suggested to provide visitors with the seasonal information they are seeking may help increase foot traffic (shown below).

2. It is advised to feature and highlight “camp” and “seasonal event” offerings. It is because a recurring search term for the Ingenium site, for example, is “camp” and a popular page for CAFM is the “Farm Event” near Easter (shown below).

2. Use “hot” or interactive media, especially on perennial favorites such as education and exhibitions pages, to increase interest to help offset drop-offs in use in December and January

2. Individual & Shared Strengths

There were key areas of strength including pages related to: visitor information, field trips, science at home, education, and exhibitions.

The visit section (all pages that start with /visit in the URI) ranked second on the level 2 page path metric for CAFM and CASM and third for Ingenium and CSTM. (shown below)

Each site also had individual strengths. For Ingenium, pages related to field trips, science at home, and (pre-pandemic) exhibition pages like art of the brick and Leonardo da Vinci were popular. For CAFM and CSTM, education pages were popular (especially the “Try this out” pages) and for CASM, the rides and simulations page drew attention.

Recommendations

1. Content about basic science, pages that include “Try This,” special events like the Farm Event, features such as rides and simulations, and various camp offerings should be as visible as possible.

2. We recommend using “hot” or interactive features (i.e. video, games, etc.) to make information about the above-mentioned dimensions as attractive to visitors as possible. 

3. Areas Needed Redesign

Site traffic is strong but there are substantial issues with bounce and exit rates across all properties.

New sessions are up overall in trend, which is positive, compared to 2020. The sites, however, are still experiencing high bounce rates.

In some cases, the bounce rate for pages is near 100%. A bounce rate of 100% may indicate that visitors found exactly what they wanted and discontinued their search. It may also indicate that visitors were drawn to the website from another source without actually wanting to go there, or that the site did not meet their needs and expectations.

The combination of time on page, bounce rate, and exit rate suggests that users are not finding the information they are looking for and are leaving.

Recommendations

1.  In the absence of more detailed user testing and qualitative research, it is difficult to ascertain why the bounce rate of the post-redesign pages is as high as it is. In general, we recommend rethinking the architecture of the site pages that focuses on an easy to use visual hierarchy.

2. Many of the pages are text-heavy, the presence of more images, graphics, videos, etc. may make the pages more attractive to visitors.

3. We recommend the use of bread crumbs during navigation to help visitors find their way back to their starting points. This may help with exit rates.

4. We recommend an exit survey or poll asking visitors whether or not they found what they were looking for and/or were satisfied with their experience. This, too, may provide insight into the bounce rates noted in our analysis.

The exit survey ideally should be short and ask: “Did you find what you were looking for?” Yes/No? With an optional short answer. This type of survey is valuable because it provides insight into what visitors hoped to find on a page why they are leaving. Such qualitative information will supplement the quantitative data provided by Google Analytics.

4. Social Media Patterns

Social media sources generate little referral traffic for the museum pages.

Organic traffic is optimal since users have a specific intent.  In general, referral from social media sources were minimal for the Ingenium properties and did not account for significant traffic.

In addition, the traffic from social media sources is associated with bounce rates above 80%. Social media can be a powerful tool for building awareness and for amplifying the museums’ messages. It is advisable to look into this performance gap to find strategies for improvement.

Recommendations

1.  Referrals from social media sites can be a way to drive traffic to the museum. We recommend adding social media share buttons and excerpts of website content, especially those for special events and exhibitions, to channels such as Facebook, Instagram, and YouTube to attract more visitors

2. We also recommend the development of museum-specific hashtags for each property to make it easier to track these metrics across various platforms.

Conclusion

Suggestions for Future Actions

1. Standardise Configurations in Google Analytics

It will be helpful for Ingenium if all dimensions of analysis are standardized across all properties in Google Analytics. For example, it is possible to examine search terms used for CAFM but not for CASM or CSTM. Being able to analyze all properties according to a standard set of metrics will help point up deeper patterns in how visitors are using these sites.

2. Conduct User Testings and Surveys

1. We encourage Ingenium to carry out user testing for video, .gif, and other interactive media to determine what visitors think and want.

2. We recommend that Ingenium carry out surveys regarding site content (most and least attractive) as well as A/B testing for any new features developed.

3. We advise Ingenium to conduct exit surveys to determine whether visitors are leaving a page because they have found the information they were seeking or are leaving because they are dissatisfied with what they have found.

3. Devise New Content Strategy

We encourage Ingenium to work with its marketing team to develop and execute a strong, focused social media strategy.

1. Users were confused over the structure of the menu items

“On the information page, I wish I could click the author, the room number," view more information related.”

Problem

Many participants were confused about the Spanish used in the labels on the navigation menu.

Recommendations

Combine the two pages Highlighted Works and 50 Selected Works into one page and rename it as 50 Highlighted Works. This action would reduce users’ options, lower  users’ cognitive load and also reduce redundancy in the menu.

Previous project

Usability Study

The Thyssen-Bornemisza Museum's Audio guide

Audio guide
in the Thyssen-Bornemisza Museum

Next project

User Research & App Design

Dayly - The Estée Lauder Companies' Skincare App

Skincare App for The Estee Lauder Companies