That's a headscratcher!

That’s a headscratcher all right!

Recently I asked my fantastic, dedicated, smart group of people who subscribe to my enewsletter, “What do you want to know more about?” I got SO MANY good responses that I could not stand to just let these answers be lost to email. I wanted to share them with you.

This time, we are answering questions about DATA and your Mailing List! A fabulous Canadian reader writes:

> Hi Mazarine –
> According to Tom Ahern, donor appeals – not prospecting – should be getting much higher renewal rates. More like 40 – 70%. And my donor appeals aren’t doing that. However, I think part of that has to do with the fact that my donor mailings are probably larger than they should be and include lapsed donors. But that’s another discussion.
> We don’t really do any donor communication between mailings except a print newsletter, and I’ve been given to understand that even newsletters should be revenue generators. We don’t have email addresses for most of our donors, so while we have the mechanism to send email communication, we just aren’t. Until we can gather that information, email is just not a tool at our disposal.
> And, due to strict privacy legislation in Canada, we are not allowed to know why people were at our healthcare nonprofit, so unless a donor tells us, we don’t know, and certainly in our patient prospecting we have no idea.
> That said, I’m hoping to do some comparisons of mailings done at similar times of year, or maybe even topic versus topic to see if one type of story has drawn better than others. But one thing I can’t seem to find consensus on, is when one does a development audit for the year, are the numbers compared totals for year over year to see if there is growth? Or mailing over mailing within the year? Or similar mailing versus similar mailing year over year? For example – what if you did a May mailing in 2011, but a June mailing in 2012? Are those comparable?
Thank you for writing!

And thank you for telling me about the constraints you’re working under. Wow, it really makes it seem like we have few rights to privacy here in the US! Ulp. I see what you mean about your privacy laws now.

3 ideas to help you get email addresses despite Canadian privacy laws:

  • Paper survey sent to past donors, “Help us get to know you!” and ask for their email. Include postage paid envelope. You don’t have to ask about what kind of medical procedure they had done, maybe just ask how their experience was, if they would like to commend any particular staff member, and then ask if you can have their email to keep them updated on your news.
  • Change your remit envelope to include a space to write their email addresses
  • Include an optional paper email signup form at your office in your check-in procedures for non-emergency visits

And yes, using a big list that includes lapsed donors would drag your averages down. Sending to donors who have given in the last 2 years would be your best bet.

So who do you ask to give in your appeals?

People who haven’t given in the last year may need a phone-a-thon to re-engage them, a week after you send the letter, use your phone-a-thon volunteers to call and say, “Hey, did you get our letter? Can we answer any questions for you about what we’re doing now, and how we’re helping people?”

I found a couple of data models from a UK fundraising researcher Paul Weigand that might help you:

What is data modeling? Go read his post and then come back.

Okay, now we’re back.

How to break down your donors into “most likely to give”

First, because your database is big but your list of recent donors is NOT so big, parse your list of donor names out to JUST people who have given in the last 18-24 months.

In the second article, Paul talks about what to do to find the people most likely to give in your database, by assigning them a “Grade” based on various factors.

For example,

  • Have they been to your open house? 1 point
  • Your gala? 2 points
  • Your golf event? 2 points
  • Have they sat on the board? 3 points
  • Have they ever volunteered for you? 3 points
  • Have they given in a previous appeal? 4 points
  • Have they given through the internet? 2 points
  • Do you know if they know someone who has been helped by your nonprofit? 5 points

If there are a number of YESES to these questions, then you know that you can assign a high “grade” to this segment of donors.
If the answer is “NO” to most questions, assign a low grade.

Then, graph this.

Here’s Paul’s second article, showing you how this graph could work on a spreadsheet.

Once you do this, you can save yourself a lot of money and effort by JUST mailing to people you feel pretty sure will give.

(If you don’t have data on who came to your events, or former volunteers or board, then do the best you can and start with the data you have. Make a spreadsheet to start if your database won’t allow you to track this.)

I did ping @PaulWeighand on Twitter and ask him if he knows statistics for Canada, or if he knows of anyone doing what he does in Canada.

Also, I found this pdf on donor behavior in Canada that might be useful to you. Here’s the link.

It doesn’t break down giving behavior by month unfortunately. But it does show across different segments of the population who gives to what. If you’re really curious, you might send an email to the creators of the report, inquiries@environicsanalytics.ca and ask if they’ll do something like Atlas of Giving or Blackbaud soon.

How to Compare if Your Mailings are Getting Better, Worse, or Stagnating?

When I was working at a small development nonprofit, they did not seem to care whether we made more money with one particular mailing, as long as our mailing dollars went up overall year over year.

I would definitely look at your numbers from let’s say, Spring 2010 and Spring 2011 and see if there was a difference. You could still compare May to June.

But realize that there’s a lot of unknown factors in this.

For example, let’s say your letter in 2010 was RADICALLY different from your letter in 2011.

  • You personalized the greeting.
  • You added quotes from patients.
  • You added a picture.
  • You added a story.
  • You sent it out at a slightly different time (May versus June), to
  • A different segment of donors.

Ideally, you could test one thing at a time to see which really made a difference, but if you do a lot of changes at once, and the mailing does better, it’s hard to say which particular change made the difference.

Do you know what I mean?

Do you have any ideas for how to measure if your mailings are working? What do YOU know about Predictive Data Modeling?


Want 43+ more tips on how to write successful appeal letters? Just go here!