Iterating Planet 4 with data & research: a 1-year snapshot
Methodology, key findings and next steps to improve Planet 4 with data
Over a year ago we “flipped the switch” and launched the first Planet 4 website (Greenpeace International). The following step for the Design team is to iterate the initial product (now deployed in 31 Greenpeace institutional sites) based on quantitative and qualitative data. Here’s what we are measuring (and validating*), how we are extracting data and a few key results to start figuring out what works and what may be reworked.
Methodology & tools
(here’s the full report or a video recap)
Using multiple approaches and tools for user research, we analysed how P4 users feel about the design in general, how they interact with the prototype and how they understand its usage.
Qualitative data
For users’ feedback we opted for both a “Feedback” button displayed on the International website and a pop-up survey (both powered by Usabilla). For usability, a series of UX tests were run, including self-guided tests recorded with Lookback, to find out how users navigate and find information on the website. Lastly, video recordings allowed deeper insights on actual website usage.
Quantitative data
Through a solid Google Tag Manager setup, we track events for behaviour and interactions, and deeply analyse them in Google Analytics. Click-through rates coming from Hotjar heatmaps and refined data (in spreadsheets) were lately integrated into DataStudio dashboards, for consolidated views.
Key findings
(here’s the full report or a video recap)
Important: we are cross-checking quantitative findings with March, April and May 2019 figures. Hotjar heatmaps are also being captured on 6 additional Planet 4 sites (Argentina, Canada, Chile, Colombia, Brasil, India and Japan), for a broader data baseline. The key findings below (and the action plan) will have to be confirmed by these benchmarks.
Design key findings
(here’s the full report or the video recap)
In general, users like the new Greenpeace website (P4) over the previous one (P3).
Negative feedback was especially around colours, the use of images or text legibility, all of which can be improved through legibility enhancements and / or new colour theme options.
Navigation key findings
(here’s the full analysis or the video recap)
Data show a generic ease of use of P4, but not specifically about its navigation. Users seem to struggle in interacting with the website, using the navigation menu mostly on the homepage, Act or Explore. On all the other pages, however, users tend to divert towards the search commands rather than Tags or categories when in need to navigate through content.
This clearly indicates that the Design team must verify these data (especially cross-check them with other P4 platforms around the world), eventually explore different UX navigation approaches and check assumptions through user research with a tree test or/and usability tests.
Findability & pages findings
(here’s the full report and the video recording)
Surveys outcomes and usability tests highlight that users sometimes struggle to find specific information. A new search engine (ElasticSearch) and a taxonomy review are recommended, to be adopted together with the above-mentioned UX improvements related to navigation.
Coming to pages, the international homepage receives high incoming web traffic, (20% of the total), also because this page is the global landing page for greenpeace.org.
Such great homepage inbound traffic, however, has a fairly poor interaction or click-through-rate (CTR), proven by a very high exit rate (46.7%) and a strong use of the Country Selector. This combination proves that a global landing page to let the users choose their country before landing on the Greenpeace International website (such as ikea.com), may be a solution.
Data shows that the Act page positively converts visits into clicks on the Take Action cards, generally confirming that users understand both page and cards’ purposes. Heatmaps show that conversion may increase by making the whole card interactive and not only the red Call To Action. Should these results be confirmed by the current benchmark with other P4 sites, the design team will test new cards design. Take action pages also fulfill their purposes with the “tasks” performing particularly well.
If “ACT” performs well, the “EXPLORE” page could definitely do better. Beside receiving low traffic (5%), the issues / campaigns split does not seem to be clear for users. The navigation explore — issue — #Tag does not seem to help users to jump from a topic to another, without going back and forth through the Explore page.
Although being accessible from any pages of the website, #Tag pages are the least visited pages. If confirmed by tests on other P4 sites, this may indicate that the P4 tag navigation system needs to be revisited.
Next steps for the prototype
1. Get to know Planet 4 audience better
The findings of the 2017 P4 user research, which shaped the first P4 version, need to be checked with quantitative data, that show a younger audience than the initial one. This might be due to the growing awareness of climate issues around the world, especially in younger generations (Gen Y & Z), who are more likely to take action than older ones. A deeper understanding of this audience is key to provide them a tailored design, content, tone of voice and user experience. A 2nd qualitative audience research has already started.
2. Run analysis and tests on other P4 websites
Planet 4 UX/UI design & data work must not be based exclusively on the prototype, but thought as a whole process across the P4 family. Running tests and analysis on local institutional websites is key to get a broader overview of which improvements the P4 product needs to create a better user experience. As said, we are capturing Hotjar heatmaps on 6 additional P4 sites, but ideally, all baseline benchmarks should be compared across multiple sites. The global P4 Analytics property will also help a lot.
3. Improve navigation (and content)
So far, quantitative and qualitative data prove that users struggle to navigate and find information on the website. Should these data be validated by the ongoing tests (see above in “Key Results”) and by a deeper navigation analysis, tests on a potential new global navigation and content on the key pages should be made and checked with a tree test or/and a usability test.
Want to see all this in a video? have a look below:
Agree? Disagree? Please feel free to send us feedback or questions, by commenting right below, tweet at #GPP4 or get in touch with the project team.
