Five Things we learned about GFW in 2018.
Evaluating the success of the things you make is essential to building websites and platforms that people love and want to use. Every year we look at Global Forest Watch (GFW), using Google Analytics, to see what we did well and how we can deliver even more value to our users. I was really excited to see that, for the third year in a row, we’ve seen more users coming to GFW, spending more time there, and interacting more deeply with the data. So I wanted to write this post to share some of my highlights from 2018 and how we’re using them to prioritise what we do in 2019. In short:
- Outreach and discoverability are key to audience growth.
- Curate content for deeper engagement.
- Give people something interesting to click on and they’ll click it.
- Always consider how people move through the site.
- Always check back that your target audience are using the features you build for them.
Outreach and discoverability are key to audience growth.
Compared to 2017, total time on site is up 34% and total number of users is up 27%. The main Global Forest Watch (GFW) site received 40,000 viewing hours in 2018: that’s around 110 hours every day.
While there was growth in audience size in almost every region, the growth in Africa is particularly exciting to see. There were 70% more users from African countries, spending nearly 200% more time on the site, in 2018 than 2017. A special mention should be made for Côte d’Ivoire, where the audience blossomed from 300 users to 1,500 users, thanks in part to the GFW team’s outreach in that country. The fact these users are spending a lot of time on the site, using a lot of the advanced features (more than people from other countries) and coming back more suggests that we’re also building something that people want to use once they discover it.
As well as the GFW team getting GFW out in front of new audiences, we also have other people taking the data out into the wider internet and using it to tell stories about their forests. There were three newspaper articles about GFW in Peru, Argentina and Brazil which attracted a relatively large audience. A few discussion board posts about fires and logging in Russia also generated discussion (and interrogation of the site) by a lot of Russian users. Designing data visualisations that people want to re-use across the internet (and making it easy to do so) continues to be an important source of audience growth.
On top of that, we’ve also created content that people are searching for in search engines. Search traffic continues to be our biggest source of users. We’ve always seen people come because they’re searching for information about forests, or they’re searching for “Global Forest Watch” because they remembered the platform exists. In the last year we’ve also seen a lot of people reading our How-to documentation, especially to learn GIS skills. We haven’t paid a great deal of attention to the optimisation of this yet, so we’re excited to look at ways to continue making content that answers peoples’ questions.
In short: if you build something that’s shareable and discoverable, and combine that with focussed outreach to the people you really care about, you should see your audience growing year on year.
Curate content for deeper engagement.
Thinking carefully about the core questions people want to answer helped us build a story around them, and provide options to dig deeper. It’s a strategy that can lead to enhanced engagement with your data, as we’ve seen from our own user analysis.
While time across the whole site increased a lot compared to 2017, one part of the site with the highest growth rate was the dashboards. Launched in February 2018, they replaced the old country pages to include content tailored to answer the key questions someone may have about the forests in a country and how they’re changing.
The new dashboards seem to be prompting deeper interrogation of the data. You can see this in a few different ways:
- The majority of pageviews (59%) on dashboards are for pages other than the default summary tab.
- Only a small portion of time on the dashboards (34%) is spent looking at the global dashboard, or a country’s summary tab.
- There are some countries where there are more views of jurisdiction dashboards than the national-level dashboard.
This implies there’s demand to dig deeper into the data available; the summary provides a good introduction leading people deeper into the data, whether that’s looking at specific sub-national jurisdictions in a country, or looking across the various tabs of more detailed data.
In short: if you structure your data in a progressive, curated storyline that addresses the reader’s key questions, you can get deeper engagement with your content.
If you want people to click, give them something interesting to click on.
If you want people to interact with your content and dig deeper into a story on their own, you have to make it interesting. This fact will never change. However, even if your content is amazing, that doesn’t guarantee that people will click on it and do that digging: especially if it’s not obvious, visible or simple enough to click.
On GFW, there was an 82% rise in the number of analyses completed in 2018, compared to 2017. This is great news, especially as this is one of the core features that can provide a lot of value to users. With an analysis completed, people can access more detailed information about the location they’re interested in — especially the trend in forest change — so they can raise the alarm or even take decisions to improve how it’s managed.
But that’s not the main thing I want to highlight here.
While it was only available for six weeks in the year (as the new map was launched in mid-November), the ability to click on any shape on the map (a country, a sub-national jurisdiction, a watershed etc) and quickly run an analysis on it was rapidly adopted by our users. In fact it was used 156,000 times in those first six weeks.
If it were being used evenly by every map user, there’d be 7.6 shapes clicked per user in that six week period between 15 November and 31 December. The good news is that it appears to be helping generate more analyses. However it’s not as simple as one click: the workflows we can observe in Analytics suggests people click several areas before actually performing an analysis.
In this case, we’ve added a small piece of functionality to fast-track users to a key feature and it’s worked really well. It’s a fairly prominent interaction (it applies to almost anything on the map) and one that’s framed to pique a user’s interest. This doesn’t work magically for every single interaction you build of course. The New York Times Data Science team observed that you should really only expect around 15% of your users to interact with buttons or settings you add to a visualisation you make. We’re seeing similar rates of usage of some of our customisation features, which by their nature have a much smaller target area on the screen and are not framed in the same way as to induce curiosity. In the year ahead we plan to use A/B tests and discussions with users to see how we can make sure they find and use them!
In short: Giving users something big and obvious to interact with that communicates what you get from clicking it can increase its use; the smaller the interaction hotspot, and the less information you give about the benefits of that click, the less use you’ll get.
But always consider flow through a site.
We also saw some potential issues with the way people navigate around different types of page on the site.
The Map is the most frequent landing page, the page you first see when you load the website. Third are the dashboards. Together, they are the landing destination for around 55% of all of the viewing sessions of GFW. In many cases, the user has searched for something and they’ve been directed to the GFW Map or Dashboard to answer their question.
This means that many people are landing straight on the data. The thing to consider, as a result, is the welcome we want to give people. When we’re designing the default views of these pages, we may need to consider adding more of the kind of onboarding content that you may expect in a homepage to give people the best possible welcome, instead of expecting the majority of people to have been through the homepage already and learned what the site’s about.
The other issue it poses is for the site architecture, and how we expect people to move around the site. In 2018 we saw that 64% of people that see the map land there, and 73% of people that see a dashboard start there. For 77% of people that make it to the dashboard, and 83% of people that make it to the map, that’s the last thing they see. Bringing these two statements together, around 45% of people visiting site only see the map or only see the dashboards in their visit to the site.
This could be a great thing: people are searching the internet for answers, they land on the GFW Map or dashboard, they get that answer quickly and they leave satisfied. We’re going to be conducting some usability tests in the next few months to see if this hypothesis is true, and if not how we can improve the site to help people see all the pages that are relevant to them. If you’re interested, get in touch.
In short: As well as looking at where people spend their time on a site, evaluate how they’re navigating through different pages; the quality of the experience most people get might be very different from the experience you first designed.
Always check back that your target audience is using the features you build for them.
The GFW team has around twelve countries where they are making a significant outreach effort, as I noted in the first section of this blog. These are countries with extensive forest cover and many organisations working to manage, monitor and conserve that forest. When we start building new features, these kinds of people are often at the forefront of our thinking.
There are three features in particular we wanted to evaluate this year: the country dashboards, country-specific data on the map, and forest atlases. Our hypothesis was that these features support domestic monitoring efforts, so would be used extensively by people in those countries.
When users from priority countries use the dashboards, they are normally looking at their own country; but they’re just a small part of the audience. Over 70% of time spent on priority countries’ dashboards is from foreign users.
On the country data it’s a more mixed picture: some country data is used mostly by its domestic audience, while others seem to be used more as a monitoring tool (so have a more foreign audience). The same is true for the National Forest Atlases: on average 57% of time on a forest atlas is from a foreign audience, though the largest single audience is users from the country of focus for that atlas.
So our hypothesis was correct — we’ve built some tools that are used to help people get an overview of forests in their country — but at the same time we need to remember about that (larger) external audience that’s also using those pages. As we add more data to these features we’ll need to make sure we’re considering both that domestic and foreign audience, to make sure we’re providing the most utility and impact.
In short: Take the chance to filter your data by different (and relevant) demographic characteristics so you can see if your features are being used by the intended audience or not.
Future focus.
In 2019 I hope we can focus on:
1. Continuing that strong growth in African countries, so we can continue the rapid growth in audience size and time on site, building tools for more forest monitors to use more regularly to guard the world’s forests.
2. Finding more ways for people to discover us, and make sure the places they’re likely to land in (pages full of data) give them an exciting and intriguing experience; enough to make them want to come back again and again.
3. Finding ways to get more people using those customisation options, so they can reach the specific pieces of data that most interest them.
If we can achieve these goals then GFW will continue to be an essential resource for anyone working to protect our forests, or managing related issues such as climate change, sustainable supply chains, and conservation of our planet’s water, soil and wildlife.
