September 2017

The current setting

The Humanitarian Data Exchange is OCHA’s open platform for sharing humanitarian data. Our goal is to make humanitarian data easy to find and use for analysis. Datasets in HDX are shared through organisations. This means that datasets contributors are required to register in the platform before they are able to share data with the humanitarian community in HDX. There is an internal process that ensures that organisations registering in HDX have a genuine humanitarian mandate before their registration is approved.

Launched in 2014, HDX started with a group of 13 organisation, mostly of them included in the platform by the HDX team. The initial collection of datasets programmatically scraped from these organisations was a mix of reference datasets including administrative boundaries and population statistics, socio-economic baseline indicators, airport locations, food prices, and refugee and IDPs statistics.

By August 2017, there were about 300 organisations registered in HDX and about 5,500 public datasets were available for public use. Out of the 285 organisations, 193 organisations have already shared at least one dataset in HDX, this is equivalent to 66% of the total, whereas for the rest we hope they will do so soon. The top five organisation sharing data on the HDX are: UNOSAT, Worldpop, UNHCR, OurAirports and UNISDR. HDX platform currently has an average of 15,000 active users per month and 79,000 pageviews per month.

Since the inception of HDX, organisations have been registering in the platform at the different rates. We have seen weeks where as many as five organisations were registered in HDX but also we have seen weeks where more than 20 organisations have been completed the registration process.

During the past years, we have seen an increase in registration and data sharing when there are new, rapid onset natural disasters. In 2016, 42 datasets were shared about Hurricane Matthew and the Haiti administrative boundaries dataset was downloaded more than 600 times over a three week period. Several organisations joined during the weeks following the natural event. Same situation was observed during the Ecuador Earthquake where 33 datasets were shared and a many other organisations joined the platform.

The table below shows the distribution of organisation joining HDX by month. Full list of organisations.

Table 1. Total number of organisations joining HDX by month. July 2014-September 2017

Table 2. Distribution of the current organisations by type (counts) as of September 2017

Our Strategy

The HDX Data Partnerships team has started a set of outreach activities to bring more key organisations into HDX and in this way more relevant datasets to the platform. The team will reach out to a series of non-governmental, public and private, military, national and international organisations. We will be using the decision makers taxonomy as our basis to map and identify relevant organisations within the humanitarian sector.

The strategy that we will follow to help us convince our new partners to register and share data in HDX is a combination of positioning our products and services visibly in front of humanitarian partners, showing the added value that open data can bring with some our our key successful stories, and stressing the benefits of collaborating with our interdisciplinary team consisted of data scientists, designers, technologists, social responsibility experts and innovator, including two data labs in Dakar, Senegal and Nairobi, Kenya.

We will explain how each of the components of this strategy will be carried about below.

1. Familiarizing with the org via desk reviews. It is indispensable to know your client before promoting the platform and seek engaging with them. Review as much as you can about the organisation; Learn about its focus area and its geographical area of action (scope). Review similar data available in HDX and review their level of technological awareness (availability of API).

1. Introducing HDX. This is the basis of any interaction with new organisation. An outreach package has been developed to support the initial contact with organisation representatives (ideally a data person who can also take decisions about data sharing) and consist of:
1. A generic one-pager about HDX.
2. A generic pitch deck for presentations and demos (sample)
3. Generic intro emails for:
1. Humanitarian partners (sample)
2. OCHA offices (sample)
3. Universities and academic partners (sample)
4. Private sector partners (sample)

This material is developed to support the introduction of HDX to the client and it is expected that the text of the generic email could be also adapted to make it specifically relevant for the organisation’s use case or to incorporate information derived from previous conversations that the team has had with the client.

1. Offering data services. Engaging in a long standing relationship with the client is key to keep the platform going. The list of data services that we could offer to our future and current partners includes:

• Interactive visualizations (see for example WFP Food prices visualization)
• Data skills trainings (see for example the Data Literacy training in Senegal)
• Data cleaning
• Data hunting
• HXL-ation of datasets
• Data preparation

A detailed list of services that the team can offer remotely or on-site can be found here.

1. Identification of data person. Once an initial contact has been done, it is important to identify a data focal point who will be supporting the data ingestion into HDX. Failing to do so could slow down the process.
2. Explain different ways of sharing data. Review these two blogs:
1. HDX connect
2. Three ways of sharing

Remember to promote the sharing ways in the following order, as stated in the blog:

“Public”:Data shared publicly is accessible to all users of the HDX platform, whether or not they are registered. All public data must be shared under an appropriate license (see all licence options here). Select the ‘public’ setting in the metadata field when uploading the data.
“Private”: Organisations can share data privately with their members. The administrator of each organisation controls who can become a member. Select the ‘private’ setting in the metadata field when uploading the data.
“By Request”: Organisations share the metadata of a dataset and grant access to the full data when requested by a registered user.