About the Client

Our client, an NGO with Tens of thousands of volunteers, donors and supporters who advance the vision of our client to grant the wish of every child diagnosed with a critical illness.

Business Scenario

Non-profit’s have huge datasets that they can exploit to develop statistical models that can help them optimize fundraising. Non-profit organizations need funds to fulfil their missions but also need to prove the results of their work in order to attract donors. When resources are limited, using your time efficiently and effectively is even more important. The client wanted to empower or use data science to make a bigger impact. “Is there anything that can help us make sure that we’re putting our time, effort, money and energy into the right channel?” they asked us.

Business Requirement

• Identify and segregate donors based on various factors
• Integrate large set of Data from different regions ,chapters etc
• Drive the marketing and fundraising efforts more effectively.
• Discover relationships that can help to develop specific incentives.
• Accurately measure the performance of their activities, and help them tailor their processes towards better results.
• Real-time tracking during crisis and optimizing relief efforts.

Challenges

Data Quality – user input errors, duplicate data and incorrect data linking was the first roadblock. To separate wheat from the chaff was our primary challenge.

Security – Keeping that vast lake of data secure is another big challenge. Specific challenges include:

• User authentication for every team and team member accessing the data.
• Restricting access based on a user’s need.
• Recording data access histories and meeting other compliance regulations
• Proper use of encryption on data in-transit and at rest.

Large Data management of unstructured data – Large growing data from different sources were available and integrating the same was a big challenge .The data was available form different internal systems . Most of the data was available as SQL but in different formats . Some chapters where using XLS format also . Integrating this data to a uniform format was critical

Solution

Power BI and Python were the obvious choice to visualise the NGO data on a KPI based dashboard. A team of 3 data scientists successfully deployed the solution in a span of 2 months.

We combined multiple sources of data to analyse all donors over multiple years. The data from different sources were collected in an ETL tool (Talend) which helped in Data warehousing and cleaning the unstructured data. Upon analysing donors’ responses, we developed distinct donor segments, providing a series of general and segment-specific opportunities to improve the organization’s marketing and fundraising effectiveness.

Results

The solution categorizes donors based on the denominations they usually donate and the mode of communication they usually respond to. Based on that they customized communication for those groups, and this helped to increase rate of retention of donors by 6-7% in last fiscal.

The data from different chapters and regions were integrated and analysis was given with different views and filters. They were also empowered to allocate funds based on insights from data thus use their resources optimally.

The campaign effectiveness report helped them identify the ideal time, place and channel of running an effective campaign for the best fundraising results.