By now, hopefully, you have seen the benefits of segmentation and the value of RFM as a preliminary tool.  But you have probably intuited a few of the flaws in it already:

• The distinction between recency groups is sometimes artificial. Let’s say two people give $20 per year, one at the end of January and the other at the beginning of February.  The former would be at 13 months for a January mailing; the latter at 12 months.  A few days difference could mean that one gets mailed and the other didn’t.  (In fact, I’d love to test a segmentation out to 13 months instead of the standard 12 because of this; please comment or email me if you’ve done this, so I can share and illuminate myself).
• The distinction between frequency groups is artificial. Your multidonors range from people who donate once a year over a series of years to people who donate to literally every communication they get.  RFM analysis gives these two people the same number of communications.
• The distinction between monetary value is artificial. You probably saw this one coming, given the first two.  Which donor would you prefer – a donor who donates $10 ten times per year or a donor who donates $50 once a year?  RFM prefers the latter; I’m guessing you would prefer the former.
• It carries literally no other information. The $25 dollar donation of today could be from someone who clipped coupons to raise the money to donate or Bill Gates divesting himself of what he found in one cushion of his couch.  What a person donates is an indicator of capacity, but it’s a blunt tool when a scalpel is possible.
• You probably noticed I was talking only about the mail in examples. RFM doesn’t look at channels of donation, nor at the sensitivity of people to those channels.

Let’s address this last one first, as we look for ways to customize RFM more than using it alone.

Separate RFMs by channel. When you do your telemarketing list pull, you will almost certainly want to call deeper into your file to people who have made a previous telemarketing gift than those who you are trying to make their first gift on the phone.  Same thing in the mail: you may be willing to mail multidonors out to 48 months $50-99.999 for people who have given through the mail, but only to 12 months or 18 months $50-99.999 for people who have only given online or on the phone.  Separate RFMs by channel will help you make these determinations.  Remember that you are looking to make sure that you have a strong multi- or omni-channel program, but that doesn’t mean you have to be agnostic as to channel of origin or preference.

Cadence analysis.  This goes to your $10 10x versus $50 1x donor example above.  You have to figure out who likes/needs multiple communications to make a gift or gifts and who doesn’t.  Some ways you can do this:

• Look at how many types of offers you have going to a person in rotation; hammering the same key on a piano over and over isn’t music, nor will the same approach be music to your donor’s ears.
• Take a look at how many times you mail someone in acquisition. Sometimes chronic non-responders need a different offer (as with rotating your offers above), but you can also see whether someone after seven or 17 or 27 times of getting a message from you means they almost certainly won’t donate and you can bless and release.  The same holds true for your lapsed donors as well.
• Test consistent communication versus no communication versus resting periods to see what happens when you try different cadences.
• Look at frequency of gifts within a certain period of time, rather than ever. You can see dramatic results sometimes by decreasing your mailings to your one-time-per-year donors versus your multi-per-year donors.

Mission area: I mentioned this when discussing customization types, but tailoring your communications toward the area of your donor’s interest is also a legitimate segmentation target.  Why would you send that advocacy alert to people who care only about your work in schools or your calendar with pictures of dogs to your car people?

Income: If you don’t want to ask that millionaire for $17, a wealth append can get you the information you need to customize your ask strings and communications strategy.

Location: Zip codes are the poor person’s income append and can be free-ish, so that’s a potential win.  More often, though, you can use zip code modeling to breathe life into underperforming acquisition lists.  Simply find your top X percent of zip codes from your current file and ask for just those zip codes from the rental list.  It will be slightly more expensive to rent per name this way, but it can provide a 20%+ lift in response rates and/or average gifts.  This is also a way to test new lists while minimizing risk.  One caution: you don’t want to do this for all lists or you risk self-limiting your acquisition strategy.  If a list works well for you across zip codes, use the whole thing – that way you give yourself a chance to be wrong about a zip code long-term.

Demographics: Some nonprofits find that different messages work better for men versus women.  Age can also be helpful, as you work to avoid sending your one millennial donor your planned giving brochure (there’s optimistic and there’s delusional…).  With demographics, backtesting makes initial sense, where you see how people of different demographics responded to previous appeals and messages, then use that data to define your strategy.

Previous responsiveness: It sounds obvious, but it’s ignored by CRM: if someone like getting your calendar three years ago, they may like getting it again.  Replace “calendar” with member card, action alert, survey, etc., and you have the makings of a profitable add on to your usual list mix.

Those are some of the things you can add to spice up RFM.

I said at the beginning that this was intermediate segmentation.  Advanced segmentation is modeling.  The hacks above will help get you many of the benefits of modeling at a fraction of the cost, but it won’t get you all of the benefits, so definitely leave yourself open to building smarter and smarter donor modeling solutions.

Any other segmentation recommendations you’ve seen work?  Please leave them in the comments.

From yesterday, we had a sample membership mailing with these results from last year, and we were going to cut any segment that didn’t get $.33 per piece or more, since that is what the piece costs.

To do so would actually cost us dearly.  As we discussed this week, there are more goals than just the immediate gaining of revenue.  Looking at mail pieces alone with a “here’s a piece; here’s another piece” mentality can ignore what other things a mailing can do for you.  Let’s take a look at this chart of historic gross per piece by segment with two ideas in mind: groups we want to invest in and testing opportunities.

Take a look at our $50-99.99 37-48 single donors.  They have an anemic response rate of .3% and an average gift of $80, so they would only gross $.24 per piece to mail to (and thus lose $.09 for each piece sent out).  Should you mail these donors?

There are some organizations that would say no – they think that every segment should net positively in a donor mailing or that they should only do no net cost acquisition.  I’ll try to demonstrate why I think these people are wrong.

With a response rate of .3%, it would take 333 1/3 pieces to generate one donor.  At a net loss of $.09 per piece, that’s a cost of $30 to acquire that donor.  Chances are that that is higher than you are spending in your acquisition mailings to get new donors.

But you aren’t acquiring just person off the street.  You are getting someone who then slots into the 0-3 $50-99.99 M segment for the next mailing.  You can see if the person got this mailing again for their next one, they would be predicted to gross $3.62 per piece sent to them or people like them, which is very nice.

Let’s run the numbers assuming that their average gift is $50, your retention rate per year for lapsed reactivated donors is 50%, your retention rate per year for multi-year donors is 70%, and it costs you approximately $10 to mail your $50-99.99 donors for one year.  To make the math easy, we’ll assume only one donation per year (it should higher) and we’ll assume that any donation is worth a net of $40 knowing the mailing costs (in reality, you would want to look at both the possibility that someone will give multiple times per year and that you will have to mail someone even when they don’t give).

This works out to:

.5 * 40 + .5 * .7 * 40 + .5 * .7^2 * 40 + .5 * .7^3 * 40 … .5 * .7^n * 40

(Can you see why I simplified the math?)

What this basically says is that there’s a 50-50 chance of getting any future gift from this person and they have a 30% chance of lapsing every year thereafter.  We aren’t using a discount rate because interest rates now are so low.

To simplify, it’s $20 + $14 + $9.80 + $6.68 and so on.  A bit of high school calculus later and this donor will likely return an average of $66.67 to your organization.  All for the cost of $30.

If you had a magic box where you could put in $30 and didn’t know what it would give you back, but knew the return would average over $60, you’d put money in.  I myself would ask if I could put in more than $30 to speed things up, like asking the genie for more wishes.

See the full comic and other fun stuff here

In general, your multi donors are going to be far better donors.  However, you need to communicate to single time donors in order to get those multi donors.  You also need to talk to those people whose last donation has been a while to renew them for future support.

The corollary to this is that you shouldn’t just look at this segmentation and see what to cut; you should also be adding back in.  Looking at these gross revenues per donor, you are probably (hopefully) wondering why you wouldn’t want to mail 7-12 month single donors of $100+, or deeper into your $1000+ donors, or more.  These are all correct thoughts.  Looking farther back into your pieces, you might see that someone has made the previous mistake – they looked at a small sample size of a $1000+ mailing, found that no one responded, and cut the segment.

Thinking further about this, you can see that perhaps the $1000+ donor shouldn’t get this piece, but they probably should be communicated with.  These are your best and best potential donors and there probably is a way to increase their value more so than not communicating with them.

Similarly, you’d love to renew those $15-$19.99 13-24 month donors, but this also isn’t the way to do it.  Now we are going to break out of yes/no segmentation and into using segmentation to create differentiated communications.  For simplicity sake again, we’ll assume that we have four treatments we are going to try:

• This mail piece, plus a pre and post email, for our “normal” donor segment (red)
• This mail piece with “lapsed” language, plus a pre and post email, for our lapsed segments (green)
• A high-touch invitation-style mailing to higher-value donors with first-class postage to invest in getting their gift (with email and higher-touch telemarketing as well) (blue)
• A prerecorded outbound voice mail campaign, coupled with an email ask to less valuable or less likely donors to attempt to renew them without high marginal costs. (yellow)

I’ve oversimplified here.  With the high-touch piece, we’d almost certainly want to test borderline segments part with the high-touch and part with the control to see if the additional investment is worthwhile.  You’d also want to test what segments telemarketing works best with.  And so on.  But for first steps, it’s directionally correct.

So let’s color in the mail plan with these four layers.

Note that you are able to contact more people with more appropriate language with this strategy.  Segments that a pure net perspective would have ignored are renewed in this new model and our most valuable donors are treated that way.

But that’s still just RFM with some embroidery on top.  It’s a fine model, but there’s more that can be done.  We will go beyond RFM tomorrow to add a few other pointers tomorrow.

Over 80 percent of nonprofit marketers know the term “communication segmenting” according to a Bloomerang survey.  Over 60 percent say they segment their files.  This means that almost 20 percent of them don’t know what segmentation is and 20 percent more know about segmenting, but don’t.

Let’s see if we can reduce those numbers.

Segmentation, as you can see from my last two posts, really is for everyone, regardless of the size of your file. We’ve talked about, obliquely, two different types of segmentation – yes/no segmentation (will someone be communicated with or not) and segmentation to help create versions or customizations.

RFM analysis can help with both of these.  It stands for:

• Recency: Usually used as how recent was someone’s last donation to the organization. You may occasionally also look at their last interaction with the organization, but we can put that aside for now.  This is perhaps the primary driver of segmentation and, if there is an answer to the “how many people do we have on file?” question, it’s when it phrased as “how many people have donated to us in the past two years” or the like – with a time horizon and action attached to it.  Do yourself a favor and add “recency” to your Microsoft Word dictionary; you are going to be using it a lot and Microsoft Word doesn’t know that word exists.
• Frequency: We go from the most often used to the most often ignored – how many times someone has given to the organization (or interacted with in non-donor contexts). This is often simplified to single v multi, as this dyad makes it easier to plan your communication.
• Monetary value. This is usually measured by another TLA (three-letter acronym): HPC or highest previous contribution – what’s been the highest amount the person has made in any one gift. This is an area of some debate, as if someone makes gifts of $20, $20, $20, $20, $500, $20, $20, $20, and $20, it is fairly predictable that they are probably better grouped with the $20-$24.99 donors than the $500-$999.99 donors and their ask strings changes along with this (more on ask strings here).  One solution is to use a formula like 2/3 of their HPC + 1/3 of their most recent contribution or half of their HPC + half of the average of their last five contributions.  But this is something worth testing how it works best for your file.

Let’s do a yes/no segmentation with RFM.  We have a membership mail piece that has historically very well with a number of segments.  It costs $.33 to mail and you are looking for segments that net.  Here are what your RFM response rate, average gift, and gross revenue per piece would look like from last year.

Response rate:

Average gift:

Gross per piece:

(these are intentionally realistic, but false, data)

To explain, the first numbers are the months (so 0-3 is someone who last gave a gift in the past three months), S & M stand for single and multi (get your mind out of the gutter), and the dollar amounts across the top are that person’s highest previous contribution.

Looking at this, to maximize net revenue, we can cut some of the segments to lower HPC groups and to one-time donors.  Anything under $.33 per piece isn’t going to net us money.

This is a decent baseline that answers the question “I have this mail piece going out in February; to whom should it go to maximize the net revenue of the mailing?”.

However, we are going to look at it as “how do we maximize the value of this donor by treating them appropriately?” and layer in some treatment segmentation tomorrow.