By Cannon Brooke, Director of Prospect Research, University Advancement, Chapman University
Collecting, storing and analyzing donor data is now simpler than ever due to technological advances in our field. Our lives as prospect researchers have been getting easier. There is a never-ending list of modeling tutorials and data analytic webinars to check out. The last 10 years have been especially exciting for data wonks, mid-career researchers and even freshly minted prospect researchers to learn data analysis skills — from predictive modeling, to affinity and engagement scores to help find new donors.
While we have an abundance of technical tutorials, there is little discussion about what to do with or how to present the data when we receive it.
What follows are recommendations for choosing the most important variables from a wealth screen, efficiently presenting the results to non-technical colleagues, potential landmines and ways to optimize your presentation. My goal is to break down the wealth screen and help you navigate these variables without having an aneurysm. I will argue that the amount of data is less important than the quality, meaning and what you do with it. Hopefully after reading this, your data presentation will transform from a frumpy pair of Saturday morning sweatpants to a sophisticated pair of pleated trousers.
What Is Missing
Before we begin, let me be unequivocally clear: I am not bashing data science. My concern is that there is too much emphasis on the technical side and not enough on the human side. I am providing suggestions for those who do not have the means to buy the latest tools or the ability to finance a data science team.
There has been so much focus on how to slice, dice, dedupe, machine learn and write SQL queries that we need to collectively reassess what exactly we are trying to do. Data is essential, but we need to talk about the art of presentation and persuasion. It is sad but true that your analysis is wasted if your fundraisers do not take action on it.
What does a handful of easily digestible data points actually look like? How are you presenting the gems you uncovered? Did you use loaded jargon, glittery glamorizations, then send the fundraiser or executive a bloated Excel spreadsheet, hoping the AI wizard behind the curtain could sell your analysis?
If you answered yes, you shouldn’t be surprised that your prospects are sitting in development purgatory. Your research shop cannot afford being a bodega hawking random data points to frontline fundraisers.
Some Things to Think About and Data Wrangling
We capture so much data — and with that, there is the potential of ignoring, dismissing and even misleading. The first step is to recognize the cognitive banana peels many of us slip on to improve our analysis. Here I’ll focus on two I frequently see in development and prospect research, which go hand-in-hand.
Clustering bias: This phenomenon observes patterns that are chance results as indicative of a trend. In other words, clustering bias rears its ugly head when you start seeing a trend in random events in your data that occur in clusters. Save the pattern recognition for “A Beautiful Mind” — if you see a correlation between zip codes, middle names and gifts given to the local Food Bank, take a break.
Information bias: This cognitive bias is notorious with new researchers and others who hold on to the belief that long-winded bios are “still a thing.” (Hint: they’re not.) It occurs when you continue to research and add information that does not affect action. It leads to the delusion that more information reveals better results. How many times have you wanted to add all the extra data points you found? You have wealth screen capacity estimations from three different vendors, an engagement score, an affinity score, the predictive model chi-square results, RFM and statistical significance. Did you include them all?
Unfortunately, superfluous data points are worthless to most executives and frontline fundraisers. You must be strategic with this information and present only salient points. I have often found that the lack of movement from prospects is caused when fundraisers do not have the right information.
Which Data Points to Use
I will be tipping some sacred cows here.
There are a ton of variables you can look at, but I recommend sticking to five or less when presenting to fundraisers. Don’t just use estimated giving capacity. We spend too much time worrying about it. If you focus on estimated giving capacity alone, you’re going to lose good potentials and most likely highlight false positives.
Another thing to consider: How are you going to use the wealth screen? Knowing how your organization will use the screen will make all the difference when choosing variables. For example, if my shop is a higher education institution and I’m screening a list of recently deposited students, I wouldn’t weigh RFM high (or use that column). I’m looking for new prospects and 80% of the time RFM provides leads you already know.
The following are recommended data points from a DonorSearch wealth screen, what we use in our shop.* For this example, I’m also trying to discover potentials who are not on our radar. When you get your wealth screen back, it can include up to 122 different columns, which is too much for any non-data wonk to digest.
For most donor discovery projects, stick to:
- Largest gift high: This is a range and you’ll need to verify it. There is no need to add the largest gift low column since you’re trying to identify potentials who give to other institutions.
- Areas of interest: Depending on your institution, pay attention to those who give to similar causes. Pay attention to higher education, healthcare and arts counts to scoop up potential donors that support your cause.
- Foundation: This is critical for identifying people who sit on a board of a foundation and are philanthropically inclined. This column is a Yes/No/Maybe flag. Sort accordingly.
- Real estate count: Real estate is one of the only tangible hard assets we can locate, and it’s best to include. I also recommend adding states with real estate because it is great to identify snowbirds and prospects with multi-state real estate portfolios.
- Estimated giving capacity: This is the lowest weighted indicator but a good starting point for sorting purposes.
- ID: If you didn’t include this with your screen, your life is going to get messier. You need to confirm that the screened names are what you have in your database.**
Here is a good example: I am trying to discover new potentials. If my project is to find leads from my database, I would include columns like total gift amount, number of gifts, RFM score, last gift date, etc. Always determine how you’re going to use the data before beginning your project.
Presenting Your Findings
After you’ve data mined and found potentials, consider the stakeholders. If you are presenting to an executive team or frontline fundraisers, keep everything at a 30,000-foot-level in an easy-to-digest format. Failing to heed this advice will release a cavalcade of questions and you’ll spend the whole time explaining the intricacies of the data. Or worse, you'll have an audience with glazed over eyes.
Lead with all of the high-level findings, conclusions and recommendations (this is extremely important for prospect researchers), and put some analysis into it. What are your recommendations for the assignment? What fund, naming opportunity, school or college would be suitable? Are any connected to your organization through board memberships?
Don’t just regurgitate data points, numbers and factoids – be a strategist. Your stakeholders are more concerned with how to interpret your recommendations and apply them.
If the data is not presented in a way that is accessible to the decision-makers, your work is wasted. Too often our industry ignores the fact that you must be an excellent presenter and not just a number cruncher.
Formats for Presenting Your Data
There are many ways to present wealth screen results, depending on your audience and goal. Here are some of the more common (and in my opinion, best) ways to present your data.
Short-bios: Bios are the most common and requested form of research report. They are a necessary evil, but your job is to mitigate their use to prevent project creep. Stick with something simple like ID, first and last name, address, total giving, found philanthropy, and a link to their LinkedIn profile for an initial sweep. I want to emphasize the short aspect of these because if you don’t, you’ll spend a lot of time and resources here.
Wealth Screen Review: These reviews are good when you have a large number of names, and your executives want an overview of who will be at an event or trip. Apply fundamental data analysis and sort the best columns to the top. I recommend less than seven columns, but remember to add a note section to include items found outside the screen.
Here is an example:
Peer review: This gauges who knows whom from a wealth screen and will be your most straightforward data presentation. Peer reviews are best for board or executive vetting, to determine if anyone in your organization has a connection to those on the list. You only need three or four columns: ID, name, estimated giving capacity, total giving.
For peer reviews, disclaimers are critical. Let the group know the data isn't verified, and the goal of the presentation is to find connections to the names on the list. A more thorough review will come later.
A Word on Visualizations When Presenting
Visualizations can be an effective tool for presentations, but you don’t always have to include a bar graph or infographic. They can make your presentation more aesthetically pleasing, but they can also make it more distracting. Only visualize data when insights cannot be effectively conveyed without them.
Here is an example of how not to present your data.
This example illustrates nothing of value — a pivot table would be more useful. For great examples of data visualization check out “Storytelling with Data,” by Cole Nussbaumer Knaflic.
Final Thoughts and Landmines
Evaluate your data to determine which variables best tell the prospect’s story. Choose which wealth screen columns help your narrative — just don’t overdo it. It’s tempting to include all the data because it’s exciting, but it doesn’t convince the executive or frontline fundraiser to take action. Furthermore, if you send all the data to a frontline fundraiser, they can play data miner and get distracted, torpedoing what you tried to accomplish.
When data is presented effectively with the perfect balance of analysis and strategy, you will see more movement in the pipeline, and further establish your research shop as a necessity and not just a luxury.
*I do not work for DonorSearch and I’m not being compensated in any way for using the example.
**The scope of this article isn’t to demonstrate how to data wrangle and clean up wealth screens. I wrote a previous article on LinkedIn that provides a way to identify the best data and how to visualize it.