Get our Latest Posts Right in Your Inbox

Your email:

More ways to connect!

Browse by Tag

CCAH Blog: Industry Voices

Current Articles | RSS Feed RSS Feed

The $5,000 Question: Finding the Right Ask Using Predictive Data Analysis

  
  
  

We are thrilled to cross-post this great piece from our very own Dan Atherton, published on  the NTEN blog. Enjoy!

The $5,000 Question: Finding the Right Ask Using Predictive Data Analysis

describe the imageBy Daniel Atherton
Account Executive, Interactive Department, Chapman Cubine Adams + Hussey


When we ask for something a prospective donor is not prepared to give, we may not be losing much real money, but we are losing a window of opportunity to convince her to give to us.

I got an email over the holiday season from a nonprofit I’ll call HELP, which was soliciting donations for its research institute. HELP asked if I was interested in honoring loved ones by having their names inscribed on the wall of the institute.

Of course I was! This would be a touching (and easy) holiday gift idea. I scrolled down the email, looking for more information. Then I got confused – as if I’d received someone else’s mail. I’d never donated to HELP before. In fact, I’d only recently joined their list. But what was the ask?

$5,000.

I bought those loved ones an animal calendar instead.

As online fundraising explodes, taking a larger share of nonprofits’ fundraising budgets each year, many of the lessons that were learned over decades of offline fundraising have seemed less relevant. I have no cost per thousand to send out an email. The value of a test idea doesn’t hinge on how much more expensive it is. And I lose no money when a recipient doesn’t donate.

But I do lose something: an opportunity.

Direct marketing is direct marketing – the sense of one-on-one contact that we cultivate right from the opening line of our emails is critical to establishing a sense of trust with our prospective donors. And when we ask for something that a prospective donor is not prepared to give – say, a significant portion of his yearly income – we may not be losing much real money, but we are losing a window of opportunity to convince him to give to us. And with so many other organizations and email lists competing for his attention, we may not get that opportunity back.

There is a wealth of data available to us as email marketers about how our constituents interact with our messaging. This data has uses outside the boardroom. We can use it to take a more offline, value-based approach to our online marketing. We can ask them to do things we know they’ll be willing to do – and to keep from asking things we know they probably won’t be willing to do. But how?

1) Gather as much data as possible.

What’s the best kind of data? All of it. Nothing is too obtuse. Utilize the data provided by your blast mailer and CRM; utilize the data provided by appends – age appends, gender appends, even social media account appends – or modeling services.

But many organizations forget the easiest way to get a constituent’s data: ask for it.

Surveys are one of the best ways to get information about who your constituents are and what they like. Ask them questions as soon as they sign up for your list: where did you hear about us? What are you most interested in? You can also send out surveys to your email file a few times per year – tell your constituents what your upcoming priorities are, and ask which are most important to them. Ask how likely they are to donate in the upcoming months. Ask what would make them MORE likely to donate.

Not only will they answer you – post-survey donation asks are some of the most successful, least-intrusive ways to solicit a gift. People are flattered when you ask them what they think. And the ones who share the most information with you are the ones who are most willing to put their trust in you – these are the constituents who can be cultivated to become your best donors.

2) Look for patterns.

Once you have all this data, your next job is to sift through it. Figuring out what to ask can seem difficult. But here are some basic questions and categories:

Who is engaging with your emails? Is 5% of your file driving 90% of the action? Are they men or women? Old or young?

What are they most interested in? Is it one specific part of your organization’s mission? Do they prefer to sign petitions, or to support you with donations?

When do they engage? Do they prefer the morning or the afternoon? What day of the week is best? Do they donate at particular times of year, or to very specific yearly campaigns?

How much do they donate? Is your list a few donors who give large amounts, or lots of donors who give small amounts? Do donors give more to renewals or appeals?

Why do they donate? Do your donors give to institutional asks (“We are rated four stars by Charity Navigator”) or emotional appeals (“We need your contribution now to feed starving children”)? Will they give to a donation ask, or do they prefer to give after taking action?

3) Selectively target your constituents based on which asks will have maximum value.

Sometimes the answers to these questions will be applicable to your entire file – for example, data shows that it’s usually best to send out your emails early in the morning, and to avoid weekends. But often you will see your best results when you group constituents based on differing answers to these questions. Just as your organization’s offline team might send different premiums to male and female prospects, or might call telemarketing-responsive donors more often, your online file can be selectively targeted based on what they’re most likely to respond to.

Say you’ve discovered, when looking for the patterns in your data, that your nondonor segments convert at higher rates when you appeal to them on an emotional level, but your donors are more likely to give again when you tell them how efficiently their extra gifts will be used. In this case, emphasize different things when you send out email appeals to these audiences. You might use pictures of children or puppies when you email your nondonor segments -- but use a chart of how the funds from this appeal will be used when you email your donors.

Or suppose you’ve found that a large portion of your file is taking action over and over again – but never donating. Sending these constituents your standard donation appeals over and over could be unhelpful or even counterproductive. Instead, send them an appeal that tells them how grateful you are that they take action so often – and why it’s so important for them to take one more step and become a donor.

One last example: maybe you’ve found that some of your donors give at the same times every year. Send these donors a specific sustainer upgrade ask: “We know you’re there for us every year. Did you know that giving just $10 each month would help us utilize your support more evenly?”

These are just some ideas to get you started. The main thing to consider is: while production cost can be a negligible factor in email marketing, opportunity cost certainly isn’t. You don’t have many chances to break through the crowd of emails that fill our inboxes fuller each day. But every day your constituents give you information about what moves them to read your messages, take action, and donate. Listen to what they have to say. Ask them what they want. If you ignore this information, you’re just wasting their time and yours.

(And their loved ones probably don’t even want that calendar.)

We'll have more on predictive donor value metrics at the 2012 Nonprofit Technology Conference in the session, "Predictive donor value metrics: how to find out what your file will be doing tomorrow -- and beyond".

Daniel joined CCAH's Interactive Department in 2010, and specializes in online production and analysis – email and landing page coding, graphics, data analysis and benchmarking, and more – for CCAH's diverse array of women's rights, environmental, and other progressive nonprofit organizations. He maps copy and production strategy for his clients and also leads the company's efforts to innovate in the areas of email rendering and deliverability, predictive data analysis, mobile rendering and text integration, and more.

Comments

Currently, there are no comments. Be the first to post one!
Post Comment
Name
 *
Email
 *
Website (optional)
Comment
 *

Allowed tags: <a> link, <b> bold, <i> italics