How We Problem Solve, Prototype and Product-ify Data for our Users

Today data has exploded, and it’s available in more places and forms than ever — pie charts tracking websites’ visitor groups, bar graphs tracking code commits, line charts tracking steps on a fitness tracker app, star ratings for restaurants, scores for taxi drivers… We’re drowning in data, and yet it doesn’t seem to be getting any easier to use that data. Too often, the data we need is stored in massive databases and dashboards, where it takes hours to find the right data points and days to come to a decision.

As an Account Manager at SocialCops, a data intelligence company, I deal with building my fair share of data solutions and dashboards. I’ll admit, designing them can feel like the least important step, given how much time and effort it takes to first source, clean, process, merge, and analyze the data. But I’ve actually learned it’s the most important step. Without a thoughtful design process, data solutions fail to help the people they’re meant to help.

At SocialCops, we’ve done a lot of work to figure out how to design solutions that will really help our users come to the right decision quickly. One of the toughest challenges we’ve faced so far was our SDG monitoring tracker, which we built in partnership with the UN Resident Coordinator’s Office in India. Between the lack of high-quality data, complex stakeholder requirements, and intricate user flows, it took us a year to come up with the final solution. Here’s how we went about designing it, all with our users at the front of our mind.

Background: How the SDGs put data at the forefront of development

Let’s go back to the year 2000, when driverless cars were still a Jetsons’ episode and cellphones were just a fad. The United Nations launched its Millennium Development Goals (MDGs), the predecessor to today’s Sustainable Development Goals (SDGs). The MDGs were a neat set of 8 Goals and 18 Targets that 189 countries agreed to achieve by 2015.

The Goals kicked off in 2000, but few countries made any progress until 2010, and most ultimately missed the mark on achieving the MDGs. It’s no wonder, when there was hardly any data to benchmark or track their progress. Even in 2015, some countries didn’t have enough data to know if they had achieved or missed the MDG targets! As the UNDP put it, “Whatever gets measured, gets done.” Strengthening data systems — not just regionally or nationally, but also sub-nationally — became the key to not just making process, but achieving goals.

This was reflected in the Sustainable Development Goals (SDGs), 17 Goals and 169 Targets that were launched in 2016. The SDGs are not just a global framework – they’ve actually created today’s global development language. Adding new Goals on topics like climate change, urbanization and responsible consumption helped to bring these issues to the forefront of development. In addition, the ingenious way that the SDGs’ targets and indicators have been defined actually encourages cross-sectoral and cross-national development partners to work together on solving these complex issues.

The SDGs also helped to put data and all its complexities and interlinkages front and center in the development discussion:

• The 17 Goals break down into 169 Targets, which cover all sorts of benchmarks. Here are 2 examples:
  • • Quantitative: For example, Target 2.1 (“By 2030, end hunger and ensure access by all people…to safe, nutritious and sufficient food all year round.”) is measured in India by the “prevalence of underweight children below 5 years of age”.
    • Framework-oriented: For example, Target 3.A (“Strengthen the implementation of the World Health Organization Framework Convention on Tobacco Control in all countries, as appropriate”) is measured in Sri Lanka by the “tobacco use among persons aged 18-69 years”.
• Each target and indicator can be mapped to other relevant sectors. For example, stunting (children whose height is far below average for their age) isn’t identified just as a malnutrition problem, but as something that should be addressed through health, agriculture, sanitation and education initiatives.
• Goal 17 includes an entire section focused on data — specifically, how high quality, timely, reliable, disaggregated data, statistical capacity-building and new ways of measuring SDG progress are the need of the hour.

Step 1: Identifying the problem to solve

December 2016 – January 2017

The SDGs made it clear that it was time for data to play a significant role in tracking and measuring the SDGs. But how to do this was still an open question. This was especially true in India, with 15% of the world’s population, an even greater percentage of the world’s poor, and some of the world’s biggest and most complex data systems.

Coming up with an answer started with a call from the UN Resident Coordinator’s Office (RCO) in November 2016. The ask? It was vague — create a tool to help all stakeholders involved in the SDGs make decisions about the SDGs.

We started by assessing what we had to work with. First, what about India’s SDG framework? The UN had an international framework with Goals, Targets and Indicators, but each country had to create its own custom indicators, contextualized to its particular development circumstances and challenges. Back then, MoSPI had come up with its first draft of national indicators, but they were heavily inspired by the global list, so they were not very well-suited to the Indian context. Other than that, we didn’t really have any other references to guide us on what would constitute the monitoring framework for the SDGs tracker of the country.

This framework needed work, which put us at a disadvantage before we even started. But we knew we did have two things going for us — a determined partner (UN RCO India), our data intelligence platform and learnings from the MDGs (which showed that government programs had an effect on helping India achieve the poverty reduction MDGs).

So we agreed — along with the decision-makers in the UN RCO, MoSPI and the Niti Aayog — to create a dashboard to monitor the SDGs’ progress in India through data. The dashboard would report on any indicators, targets or goals across states and districts, all limited only by the availability of data. This would help the UN and Government of India assess their progress toward the SDGs.

With our problem statement sorted out, we began the uphill journey of designing the dashboard.

Step 2: Creating user stories and prototyping concepts

January – April 2017

When we looked into the problem further, making just one SDG tracker for all of our stakeholders seemed impossible. Here’s what each of them needed from the dashboard:

• MoSPI (Ministry of Statistics and Programme Implementation) would use it to monitor over 300 national indicators, which all updated at different frequencies and were scattered between PDFs, webpages, Excel documents, offline reports and more.
• NITI Aayog (National Institution for Transforming India, the Government of India’s policy think tank) would use it to guide their conversations with national programme heads and key ministries. So they needed monthly or even weekly data updates, covering 41 different national schemes, disaggregated to the village level.
• The UN RCO (Resident Coordinator’s Office) would be conducting experiments to drive progress on particular SDGs, so they would need the dashboard to flexibly incorporate new data from primary surveys, crowdsourcing and more.
• District collectors were key to connecting decisions to actual actions on the ground, so the dashboard had to engage and interest them.
• We also wanted to engage citizens with online polls about their opinions on achieving the SDGs.

To build one dashboard that would work for all these needs, we knew we had to go back to the basics of design principles. So we started by creating a slide deck with all the user stories and priorities:

We then used these user stories to create a list of pages, followed by simple hand-drawn prototypes and page objectives.

We found this exercise to be extremely useful! It helped us agree on our goals, figure out whether we were covering each stakeholders’ needs, and quickly align on the dashboard’s structure and core components. We then moved from paper to Powerpoint mockups of every dashboard view.

Building a detailed structure on Powerpoint really helped us. We were able to figure out the details that everyone had been assuming, such as colors, without too much design time.

Most people like to skip directly to Invision or other fancy prototyping tools, but we’ve found that going back to the basics is often best. Drawing things out on a piece of paper and, if needed, creating a simple powerpoint can usually handle about 70% of the design details. And working faster on simpler prototypes also reduces iterations significantly on the final prototypes. (Google Design Sprint loves working on paper for this same reason.)

Because we started with these paper and Powerpoint exercises before jumping to the designed prototype, we were able to turn around 90+ pages of Invision prototypes in one week. (Check out the live version of the Invision prototype here.) It was smooth sailing for all involved — our designers, engineers and the UN team.

When we shared this Invision prototype with our partner (UN RCO India) and other UN agencies and partners, we got a great response. We immediately knew that we had set a strong base for building a monitoring tool to help India achieve the SDGs.

Step 3: Validating the prototype with diverse users

April – November 2017

We spent the next 6 months validating our prototype with different users from around the world. We started with our key stakeholders (MoSPI and the NITI Aayog) in India, then we ran more workshops across India and around the world.

• MoSPI organized 5 regional workshops to bring India’s 29 states and 7 union territories into this initiative. We presented our prototype and learned how we should further customize it for state-level development priorities.
• We conducted two Data for Development workshops in Papua New Guinea, where we learned that the first step for a country to track their SDG progress is bringing all their data custodians into a single place and taking stock of their data sets. (Read more about how this data stocktaking exercise unfolded.)
• We presented our prototype at the UNDP APAC Regional Conference in Manila, where we learned that the priority for most countries in the region was creating a unified data repository for the SDG indicators.
• We attended the first annual Goalkeepers Conference in the U.S. (an annual progress check on SDG progress organized by the Bill & Melinda Gates Foundation), where we learned about the importance of monitoring the impact of the SDG goals globally.

During this time, the Invision prototype underwent a lot of revisions. For instance, while meeting with NITI Aayog, we realized we needed to incorporate a view on the Indian government’s ministries, which went above and beyond the other views.

All our learnings from these conversations helped us refine our prototype and solidify our roadmap for building this SDGs tracker for India. In fact, we actually began to see patterns in what other countries needed to help them track and achieve the SDGs.

This triggered another line of thought for us — what if our solution could be adopted by other countries too?

Step 4: Developing and product-ifying the dashboard

November 2017 – present

We worked closely with our partners to complete the first working version of the tracker for India by March 2018. The dashboard, powered by our data repository, currently visualizes 202 of the 306 national indicators from over 100 public data sources.

Even as we were building the dashboard, we were still refining our design based on user feedback. A great example is benchmarks.

In the initial dashboard, when there were SDG benchmarks for an indicator (such as Target 3.1, which says that Maternal Mortality Ratio should be reduced to less than 70 per 100,000 live births), we compared India’s progress to that benchmark. During one of our feedback sessions, our UN RCO India colleagues pointed out that, in the absence of benchmarks, we could use indicators for higher geographies as a benchmark. So for states, the all-India value could be a benchmark, and for districts, the respective state values can be a benchmark.

Scalability is in our DNA, so we didn’t stop here with a dashboard for India. We wanted to think bigger. Our talks with SDG stakeholders had made us realize that every country has to monitor its progress toward the SDGs. This means that they’ll have to create an SDG data repository and visual tracker that’ll run for the next 12 years, rather than producing 12 one-time annual reports. This is critical for developing countries, which have the most to gain from achieving the SDGs.

This led us to go further and develop an SDG solution, in partnership with the UN RCO in India. It was built to help countries and organizations take a data-driven approach to tracking and achieving the SDGs by 2030.

Expanding from a single dashboard to a universal SDG solution meant we had to add greater speed and flexibility. We automated our build processes, which helped us go from developing one Goal per month to developing 17 Goals in 3 months. We also added the flexibility to customize the dashboard by adding partner logos, background images, new interlinkages, and more. Also, to empower organizations to manage their dashboards without us, we gave an option to take a private cloud transfer and make changes to the dashboard using open APIs.

Moving beyond India… Sri Lanka, here we come!

One of the early adopters of our new SDGs solution was Sri Lanka. Citra Lab and the UNDP in Sri Lanka partnered with us, kicking off the initiative with a proof-of-concept on Goal 3. Since we had built this as a solution, we were able to turn around the first version of the dashboard with 44 indicators for Goal 3 in just 2 weeks!

Citra Lab and the UNDP Sri Lanka customized the dashboard, and they recently took it live! Go ahead and explore it here.

We’re big fans of the Sustainable Development Goals. Global efforts to achieve the SDGs has opened up avenues to not just improve people’s quality of life, but to also strive for technological advancements, build robust data systems and bring together diverse stakeholders — everyone from the planners and the providers to the reapers and the citizens.

Working with all these stakeholders to turn 12 annual reports into a single dashboard has definitely been a challenge, but it has also been an incredible opportunity. Designing something so complex for so many users was such a learning experience — one that taught us so much about how to go from design to dashboard without losing sight of the people who will end up using it.

This blog was co-authored by Anjori Pasricha and Christine Garcia.