CoolData blog

18 January 2011

Time on the call and how much the alum pledges

Filed under: Annual Giving, John Sammis, Peter Wylie, Phonathon — Tags: — kevinmacdonell @ 10:24 am

By Peter B. Wylie, John Sammis, and Kevin MacDonell

(Download a PDF version of this paper here: Time on the call-Wylie-Sammis-MacDonell-Jan2011)

It’s a Wednesday evening and you’re in the call center watching your students on the phone with alums. Young people talking to older people who went to the same school they did. Lots of drama. Smiles. Frowns. Glee. Frustration. Side conversations between callers that you hope your alums will never overhear. Maybe you’re looking forward to going home and relaxing, but you’re not bored. Too much energy in the room for that.

One of the things you notice, as you so often have, is the length of the calls. Some end quickly. Some go on for awhile. And some last a goodly amount of time, maybe longer than you’re comfortable with. Perhaps you ask yourself, “Is all that time worth the effort in terms of getting alums to pledge and pledge a lot?”

As much as we’d like to, we can’t offer a definitive answer to that question. However … we can offer some findings we think are intriguing.  Admittedly, the data we have is from only one school. There’s no way we can responsibly generalize these findings to other schools. But we want to put them out there for you to think about and (we hope) test out at your own institution.

Here’s the flow of the paper:

  • The basic data we looked at
  • The questions we asked
  • The answers we found
  • Some of the implications

The Basic Data We Looked At

We were quickly reminded of something when we launched this study: If you run a call center, you have oodles of electronically stored data at your disposal. Of course, given the pressures and constraints of your job, you’re not going to have the time and leisure to forage through all that data and analyze it. We appreciate that. Nonetheless, the data is there; it’s hanging around waiting for the day when somebody can really dig into it.

This data came from a university with about 100,000 living alumni, and was collected during one fall term of calling by 26 student employees working in evening shifts of 8 to 12 people each.

In this study, we explored a tiny portion of the data that was available to us:

  • The results of the last call made to 4,785 alumni: “ALREADY PLEDGED,” “DO NOT CALL,” “NO PLEDGE,” “REMOVE FROM LIST,” “SPECIFIED PLEDGE” (including the dollar amount), and “UNSPECIFIED PLEDGE.”
  • The time (in seconds) spent on each of these 4,785 calls
  • The amount of the last gift (if there had been one) made by the alum

The Questions We Asked

These are the basic questions we tried to answer:

  • How much time was spent on these almost 5,000 calls?
  • What was the relationship between time spent on these calls and dollars pledged?
  • What was the relationship between time spent on these calls and pledge rate?
  • What was the relationship between time spent on these calls and whether or not alums gave more than their last gift?

The Answers We Found

In this section, we’ve got a bunch of tables and graphs for you to look at. We’ve done our best to make them clear and to offer our best interpretation of the findings.  If questions remain, please do not hesitate to let us know with a phone call or e-mail.

How much time was spent on the calls?

Take a look at Table 1. After you’ve had a little time to digest it, we’ll tell you what we see there.

Let’s start with the first column in the table called “Time Interval.” Notice that there are 20 of them, and that they each contain about 5% of the calls made. If you scan over to the two columns on the right (“Minimum Seconds” and “Maximum Seconds”), here’s what emerges:

  • A good half (Intervals 1-10) of the calls lasted less than three minutes (179 seconds).
  • 80% of the calls (Intervals 1-16) lasted less than five minutes (283 seconds).
  • 5% of the calls (Interval 20) lasted somewhere between seven and a half minutes (456 seconds) on up to well over half an hour (2,115 seconds).

Because we have not spent much time analyzing call center data, we simply don’t know how typical these time data are. We’d love to know how they compare to your own call center data.

What was the relationship between time spent on these calls and dollars pledged?

Okay, now things start to get a little more interesting. Take a look at Table 2 and Figure 1. (Click Fig. 1 for full-size version.)

Here’s what we see:

  • The big news from both the table and the chart is that, as the calls get longer, there is a substantial rise in the number of pledges made per attempt, as expressed in average dollars per call attempt. There are some blips, some ups and downs, but the trend is undeniable.
  • More specifically, let’s compare what came in dollar-wise from the shortest 25% of calls (Intervals 1-5) and the longest 25% of calls (Intervals 16-20). It’s $3,910 versus $62,491. Any way you do the math, that’s a big difference.

What was the relationship between time spent on these calls and pledge rate?

Look at Table 3 and Figure 2. (Click on Fig. 2 for full-size version.)

We see the same pattern here as we did with pledge dollars received. If anything, the relationship is a bit stronger. We’ll do the same comparison we did previously, except this time we’ll be looking at pledges received as opposed to dollars received. For the shortest 25% of calls (Intervals 1-5): 37 pledges. For the longest 25% of calls (Intervals 16-20): 678 pledges. This is a big, big difference.

What was the relationship between time spent on these calls and whether or not alums gave more than their last gift?

We think this is an interesting question. After all, I think we all want alums to give more this time than they gave the last time, whether we reaching out to them by a call or a letter or a site visit or whatever.

Same suggestion: Look at the table and the figure. Then we’ll tell you what we think. (Click on Fig. 3 for full-size version.)

No doubt about it. The longer the call lasted, the more likely alums were to give more this time than they had the last time.

Some of the Implications of What We Found

The first thing we’d like to say is more along the lines of a caveat than an implication. And the caveat is that none of us should fall into the trap of making a causal inference here. For example, we can’t conclude that encouraging callers to spend more time on the phone with alums who pick up the phone is going to increase pledge participation and pledge revenue. Spending more time on the phone with these folk may pay off, and it may not.

Why do we say that? Because time spent on the phone may not be the causal factor at work here. What may be making the difference are factors related to how much time the caller spends on the phone with the alum. It could be that warmer and chattier callers are better at raising money with alums who are also warm and chatty – and it’s that combination of traits that leads to longer calls, and it’s the chemistry between the two people that produces the pledge. We don’t know.

What we do know is that not analyzing at least some of that call data (that is so very much at your disposal) is a bad thing. Somebody in your shop or related to your shop should be working the heck out of that data. And you vendors who support call center data? We have to shake our finger at you. You have the resources to do this type of analysis. Let’s get the lead out!!

(Thank you to Devin T. Mathias, Consultant with Marts & Lundy, Allison Bass, currently on professional hiatus and formerly of Seattle Country Day School and The Bush School in Seattle, and Sharon R. Lonthair, Managing Director of Development and Alumni Relations, Rochester Institute of Technology, who all graciously agreed to review this paper and provided helpful advice.)



  1. Kudos – great stuff! (And thanks for the mention!)

    Comment by Devin T. Mathias — 18 January 2011 @ 5:05 pm

  2. Thank you for sharing this, I found it quite interesting. I would like to do a similar analysis on the telefund results at my university. As a further analysis, I would be interested in seeing if time spent on call compared with payment type method is correlated. I suspect there is a relationship between a longer call and making the pledge payment immediately via a credit card. This would be a moot point if the organization analyzed does not accept credit card payments.

    Comment by Kevin Riegle — 18 January 2011 @ 5:27 pm

    • Yes, it would be interesting to take apart the elements of a call and see how much unexplained variance was left over.

      Comment by kevinmacdonell — 24 January 2011 @ 3:07 pm

  3. Were segments/donor types considered in this data?

    Comment by Gloria M. Goosby — 20 January 2011 @ 3:47 pm

    • Gloria: No, the data was aggregated, so it included both donor and non-donor segments of all types. We were interested in general patterns.

      Comment by kevinmacdonell — 24 January 2011 @ 3:06 pm

  4. the data seems to prove pareto’s law based on the number of people interested…and who would lean towards talking more than to suggest it is about lengthening calls to people who really have no interest in talking…..

    Comment by jay goulart — 21 January 2011 @ 4:15 pm

  5. […] year, the three of us posted a paper based on calling data from one higher education institution (Time on the call and how much the alum pledges). You can go back and take a look at the paper, but its essence is the strong relationship we saw […]

    Pingback by Phonathon pledges and time on the call: Another look « CoolData blog — 16 August 2011 @ 11:54 am

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

Blog at

%d bloggers like this: