Artificial Intelligence is being used to help companies better target those customers that are most likely to engage with them. Targeting non-receptive customers is an expensive mistake. If we knew who was most likely to engage in campaigns, we could prospect with confidence, maximize donations per campaign and provide an excellent customer experience which would allow us to realize the full lifetime value of supporters.
Whether marketing to a house file or to prospects, targeting the wrong audience costs more than just postage and packaging. It can drastically reduce the lifetime value of existing customers which weakens your base and dilutes your marketing efforts across the board.
A reputable fundraising campaign management company in Washington, D.C. wanted to know if it was possible to accurately predict who would be the most receptive to their direct mail campaigns, thus the most likely to donate.
They have been meeting their customers’ expectations for 27 years and are experts in direct mail fundraising. However, their most critical campaign decisions were dependent on human experts that relied on a combination of traditional analytics reports and accumulated domain knowledge to decide how to test and roll out a campaign. They achieved good results most of the time, but they wanted great results all the time. They needed a system that could predict both response rates and amount so they could optimize their outreach, and their return.
Artificial Intelligence and Machine Learning can be very effective at providing insight if given enough of the right kind of data and the proper training to learn from. In the real world, however, the data is often distributed across sources, incomplete, unstructured and sometimes dynamically changing. On top of that, the company was only able to provide a small subset of their historical fundraising data in which to learn from, which presents a challenge in and of itself since it is much easier to find patterns in larger datasets.
To overcome these challenges facing traditional analytics, ConvergentAI™ has developed Swarm Intelligence, a predictive analytics solution that was designed from the ground up to analyze, learn from and find patterns in heterogeneous data that is consists of many different types of data elements including unstructured (ie text) and structured (ie values). This system does not need to see the whole picture before it can identify patterns. The algorithms start learning from whatever data is available and continue to learn as new data is presented.
The Swarm Intelligence algorithms learn from past campaigns to discover the specific patterns that led to different engagement by different donors. The system creates a model of donor interest to specific content - across all the past campaigns. Armed with this model, the system analyzes new campaigns and recommends the subset of the donors that will be most receptive. This will yield the best return rates at the lowest investment cost.
Using SmartCampaigns, the customer could send 27% less mail and still realize an 18% increase in total donation amount, resulting in an increase in donation per request of 62% -- all because we could connect, at a very granular level, the specific variables that mattered most to specific people which would motivate them to donate.
Artificial Intelligence can provide powerful insight to customer engagement if it can effectively analyze the real world data that companies deal with. With a tool like SmartCampaigns, marketers can focus their time and budget on only the very best audience that is most likely to engage with them, based on complex reasoning, in a repeatable way.