Bill: Good afternoon, everybody. Welcome to today’s Allegiance Fundraising Group webinar. My name is Bill Blinstrub. I’m a senior AE here at Allegiance Fundraising. My co-author for this webinar is Amy Houke, our media director, who is online now. She’ll be on at the end as well to help us answer any questions anyone has on this presentation.

List models, or modeled lists as they are also called, are something we’ve been using for a long time as an agency. They tend to work very well for our clients. We try to include them whenever possible in acquisition and lapsed mailers, because they almost always improve response rate. We really hope that by the end of this presentation, that we’re able to give you all a better understanding of how these models are built and also hopefully bust some of the myths that we hear from organizations about list models and how they violate data sharing agreements, how they’re difficult to build, or too expensive is another one.

A list model is a really powerful tool that many organizations use to find new donors, and it’s often much easier and cheaper to do than most organizations realize. We here at Allegiance really feel strongly that the whole modeling process is very misunderstood. And we’re really hoping that by talking to you all today, that we can really provide some information and answer as many questions as we can so that your organizations can take advantage of this really wonderful tool, the list model, or modeled list.

For simplicity’s sake I’m going to talk mostly about models for use in acquisition campaigns, but models can be used in pretty much every segment of fundraising.

First things first, what the heck is a list model? So a list model or modeled list is simply a list built using actual donor data from your organization and others. It is used to attempt to predict giving behavior. A list model uses data not guesswork, unlike, for example, testing a list from another similar organization whose donors may look like yours. A modeled list or list model uses actual hard known donor behavior data to determine who to target.

Now, yes, this does mean you need to provide your data to someone to build the model, but we will talk more about that in a bit.

A modeled list looks at the demographics of responders versus non-responders demographics like age, income, presence of children, et cetera, to see which is the most predictive of the response in the giving behavior that we’re trying to replicate with the model. In the absence or unavailability of mail files or responder files from prior campaigns, we can still model or clone your current givers who you would like more of.

Why don’t organizations use list models? With direct mail, we typically think about creative and offer first. You start with creative because it’s visual and more importantly, it’s really something that’s tangible. You can hold it; you can mark it up. You can debate with your team which looks better. But as much as we all like to talk about creative direction and offer, which are important, the data tells us that it really is the list. Who you mail is what really determines whether or not your direct mail campaign is successful. In fact, 60% of the driving factor in your mail success is based on correctly identifying the right audience to target, the right hands to put your message into. Your best creative package doesn’t work if it’s not delivered to the right person. Your best offer won’t resonate if it doesn’t get to the right person.

Getting your direct mail package into the hands of the right person is critical for the success of your program. We all know that. To do that consistently, it really helps to have some data to inform your decision. Think about the data that your organization already uses in its programs. You use data selects for direct mail.

You use it for your add gift, for your lapse targeting. You look at giving history, giving frequency, giving level, length of time on your file, original source. I could go on. Also think about the ads when you’re online, or flyers you get in your mail, most of those find you by using data specific to your behavior.

Shouldn’t you be finding your donors leveraging hard data as well? Yes, you absolutely should. This was what we believe. Problem is you don’t have data on prospects.

Many companies can build models using your data, but list cooperatives can build it using not only giving behavior of the donors to your organization, but also by using the giving behavior of donors to other organizations as well. This additional information really deepens your understanding of individuals and better targets prospects for your acquisition mailings.

As I said, you must give a co-op your donor data to join. And we recognize that this requires you to review your data sharing policy. And some of the more strict policies may not allow your participation in a cooperative. We’ll discuss in a little more detail why participation in a co-op though may not be a violation of your policy after all.

Why the need for your organization’s data? It’s because the co-op needs actual real-world data specific to your organization, and only you have this. So what exactly is a co-op and what does it do with my data? Let’s talk about what a list co-op is. List cooperatives are data repositories housing giving transaction information of all their member organizations.

Joining requires a contract and regular contributions of the specific giving transaction data of all of your members or donors. This is what we refer to as a give-it-to-get-it arrangement. In exchange for revealing your own donors giving behaviors, you get access to the giving behaviors of dozens of other organizations.

These contracts are easily exitable. Meaning if and when the co-op stops working for you, you are able to easily terminate your contract. You do not need to participate until the end of time.

And if you think about it, it’s not so different from exchanging lists with other nonprofits, which most of us already do an acquisition, anyway. Only in this case, instead of a nonprofit, you are working with a third party.

So after hearing all that talk about co-ops and your donor data, this is a really good time to talk about myth number one. And myth number one is, a list cooperative sold my data, right? That’s how they make money. That violates my organization’s privacy policy.

Let’s talk about what they do and exactly how your data is used by the co-op. So how do list cooperatives work their magic? What did they do with your donor data? List cooperatives do not sell your data or give it away. Don’t worry. Only members have access to the names from the co-op, but mailers cannot select just your donors.

The cooperative starts with basic information on everyone in the country via a national consumer database. You are not providing any names that they don’t already have. And I’ll say that again, you are not providing any names to a co-op that they don’t already have. The transaction data that members provide is simply layered up on those existing names.

You’re simply increasing the intelligence about each of your members to help them get only appropriate mail. So if you provide a name that didn’t already exist on the file, it’s not retained by the co-op and it does not get used. In our experience, very few of our clients have had data share policy so strict as to disqualify them from building a model like this.

Many of our clients in fact, use them regularly and more often than not, the modeled lists are the best performing list on acquisition campaigns. Of course, only you know your organization’s privacy policy. But if you’re interested in building a modeled list, we really recommend at least having the conversation with the appropriate parties about it. There’s no harm in having that conversation.

As we discussed on that previous slide, co-ops have a ton of data and a ton of resources and analytical tools at their disposal. They run this deep analysis of your donor data to build a model. And this brings us to myth number two, building a modeled list is expensive, I can’t afford it.

In fact, the models cost nothing to build. You pay only to rent the names as you would any other outside prospect list. And we all pay for these prospect lists as well. Are you surprised most people are. But I’ve got more good news on this. For scoring your own deep lapsed or house names, it’s even less expensive because they’re your own records being scored and not new prospects being rented, you only pay for the scoring. So if you want to increase the response rate on your deeply lapsed files, this is a great thing to talk about if you want to really go after that low hanging fruit.

Let’s take a moment here. I want to talk about ways to use a cooperative to turn not only lapsed donors, but also non-givers in your own database into current donors. Lapse models will let you slice off only those deeply lapsed donors who are likely to give again. So this translates into huge savings by mailing only those lapse donors with the high likelihood to give. How else can you leverage your own database to find new donors?

Consider this example a free library has a very large database of library card holders. And by that we mean people who visit and use the library but have never supported the library philanthropically. By looking at their current donors, a model can be built to score those most likely to engage with the library philanthropically. Also consider a zoo or a museum who has a very large file of transactional members who essentially would buy a membership with all the perks associated with but have not given an actual gift. Cooperatives clone your donors and look at the giving behavior of your transactional members to other organizations, and that allows them to rank them in terms of likelihood to become a giver so you can focus your efforts on them.

Let’s talk about another myth. Myth number three here. It’s too much work. I don’t have time to do the data pulls necessary to build a model.

In fact, how we work, we simply build the query that produces the transaction file and set a quarterly reminder to produce and post it. It really is that easy. For Allegiance users, we can even set that up as part of our staff augmentation service and do it for you. If you’re not an Allegiance user, just check with your donor database company to see what is possible with your own database platform.

This is something that can be done with other databases as well. I have done that in the past as well. Now for the initial setup, of course, you do need to pull some transaction history. That can be a little more of a lift, but honestly, it’s not that difficult. Once clients actually do it, they’re often surprised how easy it really is. We’ll tell you exactly what needs to be pulled, answer all questions, walk you through that. We always recommend quarterly data uploads. Depending on your specific situation with your organization, co-ops are always willing to accept less frequent uploads of data. So don’t let that be a deterrent.

So who builds these models? There are four primary cooperatives: Donorbase, Target, which is Blackbaud, Abacus, which is affiliated with Epsilon, and Wiland. We have used all four, all are good. We are currently working with Donorbase building models for our public media clients, but they are all good.

There are many different model options, and they can vary depending on the data partner that you choose to work with. You can do an acquisition response model. That’s a personal favorite of mine. It does exactly what it says, it identifies good prospects for acquisition direct mail, focusing on response rates. You can build a lapsed or a non-donor reactivation model. It’s a regression model. They do a straight select panel predictive of which lapsed or non-donors are most likely to give to your organization. You can do an acquisition average gift model. This is again, it’s exactly what it sounds like.

A model that aims to bring in acquisition donors with a higher initial gift, something a lot of people are looking to do. There are other models specific to the modelers themselves. For example, if you were to work with Donorbase, you could take advantage of their core affinity model. This is constructed by a direct data mining of Donorbase itself if you do a panel of prospects sharing the highest cause mission affinity to the organization’s historic donors.

So what have we really learned here today? Let me just summarize some of the key points we’re really trying to make. Co-ops are blind. They don’t sell your data. They only use your data to add information to what they already have. And they only use names that are on both your list and their existing list.

Second point, modeled lists do cost about the same as any other list, but typically drive better response rates. And three, the co-ops really do do all the work. All you have to do is provide some data. Many of our, nonprofit clients use modeled lists as a regular part of their fundraising programs.

They use them because they work, which is really the best evidence I can point to as why you should consider a list model for your program if you’re not currently doing them.

And that is our little webinar for the day, folks. Thank you. We will take questions if you would just bear with me for one second, while I open up that panel. I am new to this, so I’m going to apologize in advance. If there’s a question and I mess it up, please forgive me, but I promise you all this, we will follow up with everybody individually afterwards. I do have the list and I will be printing out everything. Amy, can you see the questions that are on here or do I need to read them to you?

Amy: How did the co-ops make money? They make money by renting the prospect files. Once you’ve had your model built and, you’re taking prospects to include in your acquisition mail campaign, you pay a rental fee for those, like you would any other list. And then secondarily, they also charge for the scoring work. If you send in your own house file, lapsed donors, transactional membership folks, people that you’d like to convert into philanthropic donors, they charge a scoring fee for that as well.

Hopefully that answers your question.

Bill: There’s a partial question on here. I’m not sure. Do you see the second question, Amy?

Amy: I struggle with sale versus rental. To me, rental implies something is returned. That is a traditional list term. Lists are always rented and never sold. Meaning you are not able to just integrate it directly into your database and begin to mail it at will.

Names are rented for one-time mailing purposes only.

I do see the public question. I think this distinction will be lost on a donor who doesn’t want her information shared.

Of course, you would not share any names if you’ve got a do not rent or do not exchange flag that you allow your donors to utilize, you obviously would suppress them from the transaction data that you send into a cooperative. That’s more than fine. It’s also more than acceptable to suppress major donors, your larger donors, perfectly acceptable.

Tracy asks, what is the pricing for renting this type of list? Pretty much in line, in fact, a little bit lower than most vertical lists and other donor list rentals. Average is between $75 and $85 per thousand per thousand names.

More questions?

Bill: These are great questions, guys. Please don’t be shy if you’ve got anything. These are very helpful. I don’t see any more questions here. Really appreciate everybody’s time today. I really appreciate it. Thank you for your questions, they were great. And I hope to hear from you all again.

Amy: Take care.

Bill: Bye bye all.