Fundraising Analytics

When Is Enough Enough?

It’s estimated that as much as 30% of annual giving occurs in December and that for some organizations this can represent as much as half of their annual budget.

So, it is a big deal. And you’re right to want to do it right. But how do you know what right is?

One question we often hear is “How many contacts/outreaches should we have?”

Proponents of asking more typically argue more revenue justifies more asks. Opponents to asking more counter with concerns about donor fatigue. 


The problem is, both may be right, a phenomenon that has been label “the frequency math effect.” An IBM study of email last holiday season concluded “During heavy volume time frames, the higher cadence of emails typically produces fewer opens and clicks per message sent but more opens and clicks in total.”

So, how might that apply to fundraisers?

The chart above summarizes the Year End results of two subsequent campaigns for one nonprofit client. This organization wanted to aggressively increase Year End revenue, so added one new email (email #5) and one resend (email 4a). (Emails 3a and 4a were re-sends of the original message to those constituents who hadn’t open it within the first 48 hours.)

Year 2 saw a 29.6% increase in total revenue, well above standard industry increases for the year. But it also saw a 29% increase in total unsubscribes and revenue per email sent dropped nearly 25%. Greater return, at a greater cost.

So, now, what should they do?

The fact is, most organizations have multiple donor development objectives: immediate return, on the one hand, and cultivation of future potential, on the other. It may not be possible to give equal weight to both objectives in every message, or even in every month. But somehow, a balance must be maintained.

Our recommendation was to slow the cadence considerably in January, and to focus solely on more engaging content. And of course, to continue to monitor open, click-through and unsubscribe rates across the file.

What are you doing?

"What's the ideal ask string?"

It's a common question.

But the most common response – "Test, to see what works for you." – doesn't provide much actionable direction. Particularly if your organization hasn't done prior testing/tracking of ask amounts.

One way to begin is to ask your donors what your ask should be.

Literally. Look at recent donor history to validate trends in your own file.

We recently wrote an appeal for a private university that had a "standard" ask of $50, $500 or $5,000.

There's a certain alliterative logic, perhaps. But, beyond "That's just the way we've done it," there didn't appear to be a firm rationale.

We asked if we could look at giving patterns of current donors (people who had given in the past five years).

Initially we focused on the most recent gift and average giving (and with a little more time and data, would have broken out first gift amounts, too). By either standard, more than 90% of gifts were $250 or less; 80% were $100 or less; and fewer than .5% (one half of one percent!) were $5,000 or more.

Which makes alliterative logic seem slightly less sound, doesn't it?

We also looked at variance, i.e., was the donor's last contribution higher or lower than the average giving amount?

This was encouraging. About 20% of the gifts were $25 or more higher than the average (about 10% were up $100 or more). Conversely, only about 10% were down $25 or more (only 4% fell by $100 or more).

In other words, more donors tended to give more on subsequent gifts … but not thousands more.

Based on these quick calculations, we suggested a couple of options to test against the current ask string.

For people who had made a past donation, we recommended a series of graduated asks (simpler to implement than individually calculated series). Each had three levels, with the lowest level just slightly above the donor's last gift. Only one of the series went as high as $5,000 … and it would be sent to less than 5% of the base.

For people with no prior giving history, we also suggested a three-level ask, but one slightly more conservative than the previous standard. The highest level, in fact, was just over $1,000 and, coincidentally, was also the entry amount for one of the school's key giving groups.

Before you ask ...

Unfortunately, we can't tell you which string won because the test was never actually implemented. The client opted to roll-out the test ask strings instead.

But that's another blog post entirely!